cocosuperintelligence 1.5.0 → 1.6.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +168 -663
- package/bin/coco-bootstrap.sh +11 -3
- package/bin/coco.js +16 -5
- package/package.json +2 -2
package/README.md
CHANGED
|
@@ -1,86 +1,125 @@
|
|
|
1
|
-
|
|
1
|
+
# CoCo Super Intelligence
|
|
2
2
|
|
|
3
|
-
|
|
3
|
+
495 personas across 13 departments, **228 skills**[^installed] and **387 commands** in the full catalog (coco main, 2026-10-05).
|
|
4
4
|
|
|
5
5
|
<picture>
|
|
6
|
-
<source media="(prefers-color-scheme: dark)" srcset="
|
|
7
|
-
<img src="
|
|
6
|
+
<source media="(prefers-color-scheme: dark)" srcset="docs/readme/hero-dark-mode.png">
|
|
7
|
+
<img src="docs/readme/hero-light.png" width="100%" alt="Many small clusters of experts converging on one decision">
|
|
8
8
|
</picture>
|
|
9
9
|
|
|
10
|
-
|
|
10
|
+
[](https://github.com/coco-research/coco/actions/workflows/ci.yml)
|
|
11
|
+
[](https://github.com/coco-research/coco/releases)
|
|
12
|
+
[](skills/)
|
|
13
|
+
[](commands/)
|
|
14
|
+
[](systems/superintelligence/)
|
|
11
15
|
|
|
12
|
-
|
|
16
|
+
Open-core: the core is [MIT](LICENSE), and Super Intelligence is [proprietary](systems/superintelligence/LICENSE). Files stay on your machine. No telemetry. A version check may contact GitHub once a day.
|
|
13
17
|
|
|
14
|
-
|
|
18
|
+
## Why it exists
|
|
15
19
|
|
|
16
|
-
|
|
20
|
+
A hard call in a coding session usually gets one model's opinion, with no name on it and no record of who disagreed. CoCo seats a cross-team board of 495 personas, each modeled on the public writing of a named expert, answering from cited public stances, and returns one pick plus the dissent. They are not those people's views or endorsement. The same install carries the skills and commands that do the work after the verdict, and writes phase state to disk so a cleared chat does not erase the plan. It is not a new model and not a new harness. It installs into the one you already run.
|
|
17
21
|
|
|
18
|
-
|
|
22
|
+
## See it
|
|
19
23
|
|
|
20
|
-
|
|
21
|
-
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
[](systems/superintelligence/)
|
|
25
|
-
[](https://github.com/coco-research/coco/actions)
|
|
24
|
+
<picture>
|
|
25
|
+
<source media="(prefers-color-scheme: dark)" srcset="docs/readme/si-decide-dark.png">
|
|
26
|
+
<img src="docs/readme/si-decide-light.png" width="100%" alt="Terminal capture of /SI-Decide: Should a 10-person startup move its public REST API to GraphQL this quarter? Sixteen personas, shown as role archetypes from five departments, give positions, react in rounds, and return a conditional no with two named dissenters.">
|
|
27
|
+
</picture>
|
|
26
28
|
|
|
27
|
-
|
|
29
|
+
*Personas are modeled on the public writing of named experts and are not those people's views or endorsement. See the [persona disclaimer](systems/superintelligence/DISCLAIMER.md). To request removal or correction of a persona, see [docs/personas/TAKEDOWN.md](docs/personas/TAKEDOWN.md).*
|
|
28
30
|
|
|
29
|
-
|
|
30
|
-
<a href="#install"><kbd> Install </kbd></a>
|
|
31
|
-
<a href="#the-board"><kbd> The Board </kbd></a>
|
|
32
|
-
<a href="#the-framework-underneath"><kbd> Framework </kbd></a>
|
|
33
|
-
<a href="#skills-catalog"><kbd> Skills </kbd></a>
|
|
34
|
-
<a href="#system-bundles"><kbd> Systems </kbd></a>
|
|
35
|
-
<a href="#see-it-run"><kbd> Demos </kbd></a>
|
|
36
|
-
</p>
|
|
31
|
+
An illustration built from a real `/SI-Decide` run (Claude Sonnet 5.5, 30 September 2026, US$0.41). Persona names are replaced with role archetypes and the output is trimmed for length; no other words are changed. The full output, with the same replacements, is in [`docs/readme/terminal/si-decide-full.md`](docs/readme/terminal/si-decide-full.md). Note what it admits: the embedding router was offline, so it fell back to keywords, and stances applied beyond their source are marked *extrapolated*.
|
|
37
32
|
|
|
38
|
-
<
|
|
33
|
+
<img src="docs/assets/dogfood-si.gif" width="720" alt="Terminal recording of a local meta_select.py Stage-A run. The embedding endpoint was unavailable, so routing takes the keyword-fallback path.">
|
|
39
34
|
|
|
40
|
-
|
|
35
|
+
*A real `meta_select.py` Stage-A run on this machine. The embedding router was offline (`Connection refused`), so this shows the keyword-fallback routing path (`method: keyword`, `degraded: true`), not embed cosine.*
|
|
41
36
|
|
|
42
|
-
|
|
43
|
-
<img src="assets/demo.gif" alt="/SI-Decide --debate: a routed expert panel argues in reacting rounds before an attributed verdict" width="720" />
|
|
44
|
-
</p>
|
|
37
|
+
## Quick start
|
|
45
38
|
|
|
46
|
-
|
|
39
|
+
Prerequisites: git, bash, Node.js 14 or newer (for `npx`), and python3 (it generates the `/SI-*` commands).
|
|
47
40
|
|
|
48
|
-
|
|
41
|
+
`npx cocosuperintelligence` installs the latest release. It clones the pinned release tag (currently `v1.6.0`, not floating `main`) into `./coco` of the current directory, then runs `install.sh`. Run it from your home folder, so the clone is `~/coco` and not a folder inside another project. On a fresh machine it took about 3 minutes, including the download. <!-- x-release-please-version -->
|
|
49
42
|
|
|
50
|
-
|
|
43
|
+
```bash
|
|
44
|
+
npx cocosuperintelligence
|
|
45
|
+
```
|
|
51
46
|
|
|
52
|
-
|
|
53
|
-
>
|
|
54
|
-
> CoCo Super Intelligence convenes a cross-team board of 495 named experts to pressure-test your hardest calls — then puts an orchestrated team of agents to work executing the verdict: parallel build waves, deterministic verification gates, and phase state that survives every `/clear`.
|
|
47
|
+
Open Claude Code and run:
|
|
55
48
|
|
|
56
|
-
|
|
49
|
+
```bash
|
|
50
|
+
/SI-Decide "Should a 10-person startup move its public REST API to GraphQL this quarter?"
|
|
51
|
+
```
|
|
57
52
|
|
|
58
|
-
|
|
53
|
+
Newest main, which is ahead of that pin:
|
|
59
54
|
|
|
60
55
|
```bash
|
|
61
|
-
git clone https://github.com/coco-research/coco.git
|
|
62
|
-
bash install.sh --systems superintelligence
|
|
56
|
+
git clone https://github.com/coco-research/coco.git ~/coco && bash ~/coco/install.sh
|
|
63
57
|
```
|
|
64
58
|
|
|
65
|
-
|
|
59
|
+
Adapter flags and bundle selection: [`docs/install.md`](docs/install.md).
|
|
60
|
+
|
|
61
|
+
## What it does
|
|
62
|
+
|
|
63
|
+
- Get one clear pick on a hard call, with the disagreement named and each stance cited.
|
|
64
|
+
- Ask in one line. `/SI-Decide` seats 16-32 of 495 personas across 13 departments. Add `--debate` for reaction rounds.
|
|
65
|
+
- Keep going after `/clear`. Decisions, phase state, and progress stay on disk.
|
|
66
|
+
- Install into Claude Code, Cursor, Codex, VS Code, or another adapter. Your model and your keys stay yours.
|
|
67
|
+
- Carry the verdict into the work: planning, review, verification, PRDs, and a pull request, with gates that have real exit codes.
|
|
68
|
+
- Stay local. CoCo adds no analytics. Set `COCO_NO_UPDATE_CHECK=1` to skip the daily GitHub version check.
|
|
69
|
+
|
|
70
|
+
## How it works
|
|
66
71
|
|
|
67
|
-
|
|
72
|
+
<picture>
|
|
73
|
+
<source media="(prefers-color-scheme: dark)" srcset="docs/readme/how-it-works-dark.svg">
|
|
74
|
+
<img src="docs/readme/how-it-works-light.svg" width="100%" alt="From /SI-Decide through an orchestrator seating 16-32 personas to a decision, with a dashed named-dissent branch off the debate step">
|
|
75
|
+
</picture>
|
|
76
|
+
|
|
77
|
+
You type a question. The orchestrator scores 495 personas across 13 departments on domain fit, cell coverage, and productive conflict, then seats 16-32. Each seated persona is modeled on the public writing of a named expert, answers from cited public stances, and is not that person's views or endorsement. `--debate` adds reaction rounds before the tally. The decision is one pick, the dissents are named, and a stance check flags anything that does not trace to a source.
|
|
78
|
+
|
|
79
|
+
How the router scores teams: [`systems/superintelligence/README.md`](systems/superintelligence/README.md). What each editor actually receives: [`adapters/INDEX.md`](adapters/INDEX.md).
|
|
80
|
+
|
|
81
|
+
## Status and roadmap
|
|
68
82
|
|
|
69
|
-
|
|
83
|
+
Released: v1.5.0 on npm. Today `npx` installs v1.5.0. That tag already contains the 13 Super Intelligence team skill files, but its installer does not install them as front doors. That fix, and several other installer fixes, are on main.
|
|
70
84
|
|
|
71
|
-
|
|
85
|
+
The measured 190 skills / 387 commands / 34 agents ([`docs/delivered-counts.json`](docs/delivered-counts.json)) apply to main.
|
|
86
|
+
|
|
87
|
+
Known gap: most GSD skills need the upstream GSD toolkit at `~/.claude/get-shit-done` (see [`systems/gsd/README.md`](systems/gsd/README.md)).
|
|
88
|
+
|
|
89
|
+
Next three milestones:
|
|
90
|
+
|
|
91
|
+
1. v1.6.0, so `npx` installs what main already has.
|
|
92
|
+
2. Bundle the GSD runtime, so those skills work without a separate toolkit.
|
|
93
|
+
3. Installer safety: an uninstall guard, and a clearer clone location.
|
|
94
|
+
|
|
95
|
+
<details>
|
|
96
|
+
<summary><strong>Other install paths, updates, uninstall</strong></summary>
|
|
72
97
|
|
|
73
|
-
|
|
98
|
+
From a checkout you already have, `bash install.sh` auto-detects the editor. Narrow it with `--adapter cursor` or `--systems gsd,brain`. `--core-only` skips bundles. `reverse-skill` stays opt-in: `--systems reverse-skill`.
|
|
74
99
|
|
|
75
|
-
|
|
100
|
+
The remote bootstrap clones the pinned release tag (currently `v1.5.0`, not floating `main`) to `~/.coco`, prints the commit, and waits for you to type `y` before it runs `install.sh`:
|
|
101
|
+
|
|
102
|
+
```bash
|
|
103
|
+
bash <(curl -fsSL https://raw.githubusercontent.com/coco-research/coco/main/bin/coco-bootstrap.sh)
|
|
104
|
+
```
|
|
105
|
+
|
|
106
|
+
`npx cocosuperintelligence update` checks out the pinned release tag. A git clone updates with `git pull --ff-only && bash install.sh`. `bash scripts/check-update.sh` prints the version. The check contacts GitHub at most once a day and sends no telemetry.
|
|
107
|
+
|
|
108
|
+
Uninstall removes symlinks whose target is this clone, and only those. Run it from the clone root:
|
|
109
|
+
|
|
110
|
+
```bash
|
|
111
|
+
CLONE="$(pwd)"
|
|
112
|
+
find ~/.claude ~/.cursor ~/.grok ~/.copilot -type l -lname "${CLONE}/*" -delete
|
|
113
|
+
for app in "Code" "Code - Insiders" "VSCodium" "VSCodium - Insiders"; do for base in "$HOME/Library/Application Support" "${XDG_CONFIG_HOME:-$HOME/.config}"; do [ -d "$base/$app/User" ] && find "$base/$app/User" -type l -lname "${CLONE}/*" -delete; done; done
|
|
114
|
+
```
|
|
76
115
|
|
|
77
|
-
|
|
116
|
+
A contains-match on the folder name would also delete links into a path that merely contains that name (for example `coco-research` when the clone is `coco`). Hermes profiles live under `~/.hermes/profiles`; use the same `"${CLONE}/*"` prefix there. Uninstall does not remove the generated `SI*.md` command files or the rules block between `<!-- coco:rules-start -->` and `<!-- coco:rules-end -->` in `CLAUDE.md`; delete those by hand.
|
|
78
117
|
|
|
79
|
-
|
|
118
|
+
</details>
|
|
80
119
|
|
|
81
|
-
|
|
120
|
+
## The board
|
|
82
121
|
|
|
83
|
-
Super Intelligence
|
|
122
|
+
Super Intelligence summons a panel inside the session you already have open. Every contribution is from a persona modeled on the public writing of a named expert and grounded in a cited public source. It is not that person's views or endorsement.
|
|
84
123
|
|
|
85
124
|
```bash
|
|
86
125
|
/SI-Decide "Should we migrate our database to pgvector?"
|
|
@@ -107,685 +146,151 @@ Super Intelligence is CoCo's signature capability. It lets you summon custom, re
|
|
|
107
146
|
| **Healthcare & Life Sciences** | 25 | Health economics, genomics, clinical medicine, public health policy |
|
|
108
147
|
| **Sales / GTM / Marketing** | 23 | Product-market fit, sales ops, growth marketing, enterprise GTM |
|
|
109
148
|
|
|
110
|
-
- **342 generated slash commands.** 17 cross-team
|
|
111
|
-
- **
|
|
112
|
-
- **
|
|
113
|
-
-
|
|
114
|
-
- **`--debate` deliberation.** Panels can argue in reacting rounds before returning a verdict, surfacing named dissent instead of false consensus.
|
|
115
|
-
|
|
116
|
-
### The selection algorithm (two-stage routing + approval)
|
|
117
|
-
|
|
118
|
-
The meta-orchestrator uses a staged algorithm so it never has to load the entire persona filesystem at runtime:
|
|
119
|
-
|
|
120
|
-
1. **Stage A — team & cell routing (cheap).** Reads the meta-registry (`systems/superintelligence/registry.json`) and scans only team descriptions and cell definitions, scoring relevance by keyword and domain overlap to select the top 1–4 teams. If one team dominates, it delegates to that team's own orchestrator.
|
|
121
|
-
2. **Stage B — scoped persona selection.** Loads compact persona records (slugs, cells, domains, stances, conflict mappings) for only the selected teams, allocates the 16–32-person budget proportionally, runs the greedy scorer (`0.40·domain + 0.30·cell-coverage + 0.30·conflict-pairing`), and applies a cross-team tension pass to pair opposing viewpoints (e.g. security vs. growth).
|
|
122
|
-
3. **Stage C — approval & execution.** Presents the roster grouped by team with a one-line rationale. The action verb then consumes the approved roster and returns output with full line-by-line attribution (e.g. "Aswath Damodaran (Finance): …").
|
|
123
|
-
|
|
124
|
-
### Persona build & validation pipeline
|
|
125
|
-
|
|
126
|
-
The 495-persona roster was compiled with a systematic multi-tier workflow. Candidate generation was evaluated across local Qwen (via LM Studio), a hosted small model, and Gemini Flash, but **Claude research agents proved the quality winner** for resolving historical data and citing verifiable signal. Personas are checked by an advisory validation gate (`validate_persona.py`) that enforces five things: every required frontmatter field is present; at least 4 cited URLs resolve live (non-404); every `public_stance` carries an `evidence_url`, so no stance is uncited; at least 2 recent signals within 12 months (or 2 persistent signals for historical archetypes); and any `pairs_well_with` / `productive_conflict_with` slug refers to a real roster member. Two limits are worth stating plainly: the gate confirms that a stance is *cited*, not that a quotation is authentic, and `home_team` is only checked for presence, not validated against the list of real teams. A `NEEDS-TOPUP` verdict is advisory — it does not currently affect which personas a council selects at runtime.
|
|
127
|
-
|
|
128
|
-
---
|
|
129
|
-
|
|
130
|
-
<a id="the-framework-underneath"></a>
|
|
131
|
-
|
|
132
|
-
## The Framework Underneath
|
|
133
|
-
|
|
134
|
-
The board decides. The framework ships. Three pillars turn the verdict into merged, verified code.
|
|
135
|
-
|
|
136
|
-
<table>
|
|
137
|
-
<tr>
|
|
138
|
-
<td width="33%" valign="top">
|
|
139
|
-
|
|
140
|
-
### 1. Autonomous team execution
|
|
141
|
-
`/team:ship` runs a fully coordinated build pipeline — 6 development stages, 7 hard verification gates, and 2 architecture conformance gates. Specialized agents act in their domains (Research, Architecture, Plan, Review, Build, QA). Code must clear the Test Evidence Protocol (independent re-runs, strict coverage, skip-prevention) **and land where the plan said it would** before a pull request is opened.
|
|
142
|
-
|
|
143
|
-
</td>
|
|
144
|
-
<td width="33%" valign="top">
|
|
145
|
-
|
|
146
|
-
### 2. Disk-persistent state
|
|
147
|
-
AI chat context is fragile and wipes on `/clear`. CoCo persists phase state, decisions, and progress logs to disk across resets, with atomic git commits for every successful step — so you can review history, roll back a bad design path, or resume a long-running migration from any checkpoint.
|
|
148
|
-
|
|
149
|
-
</td>
|
|
150
|
-
<td width="33%" valign="top">
|
|
151
|
-
|
|
152
|
-
### 3. Vendor-neutral portability
|
|
153
|
-
CoCo compiles its rules, templates, and agent definitions into pure Markdown and YAML frontmatter. Native IDE adapters inject the same library into Claude Code, Cursor, Codex CLI, PI-Desktop, VS Code, or any `AGENTS.md` tool — so your workflows follow you even if you switch AI editors.
|
|
154
|
-
|
|
155
|
-
</td>
|
|
156
|
-
</tr>
|
|
157
|
-
</table>
|
|
149
|
+
- **342 generated slash commands.** 17 cross-team commands (`/SI-Decide`, `/SI-Tradeoff`) plus 325 per-team commands (25 verbs × 13 teams, such as `/SI-AI-Decide`). Generated on your machine at install time. Nothing is uploaded.
|
|
150
|
+
- **98 expertise cells.** Departments are split into cells so the router can seat a precise panel. The count is the sum of the `cells` fields in [`systems/superintelligence/registry.json`](systems/superintelligence/registry.json).
|
|
151
|
+
- **Two-stage routing.** Stage A reads the meta-registry and picks the top teams. Stage B loads compact records for those teams only and seats 16-32 people with weights 0.40 domain, 0.30 cell coverage, 0.30 productive conflict, then pairs opposing views. If one team dominates, that team's own orchestrator takes the question.
|
|
152
|
+
- **`--debate`.** The panel argues in reacting rounds, then returns a verdict with named dissent instead of a false consensus.
|
|
158
153
|
|
|
159
|
-
|
|
154
|
+
The 495-persona roster was compiled with a systematic multi-tier workflow. Candidate generation was evaluated across local Qwen (via LM Studio), a hosted small model, and Gemini Flash, but **Claude research agents proved the quality winner** for resolving historical data and citing verifiable signal. Personas are checked by an advisory validation gate (`validate_persona.py`) that enforces five things: every required frontmatter field is present; at least 4 cited URLs resolve live (non-404); every `public_stance` carries an `evidence_url`, so no stance is uncited; at least 2 recent signals within 12 months (or 2 persistent signals for historical archetypes); and any `pairs_well_with` / `productive_conflict_with` slug refers to a real roster member. Two limits are worth stating plainly: the gate confirms that a stance is *cited*, not that a quotation is authentic, and `home_team` is only checked for presence, not validated against the list of real teams. A `NEEDS-TOPUP` verdict is advisory. It does not currently affect which personas a council selects at runtime.
|
|
160
155
|
|
|
161
|
-
##
|
|
156
|
+
## Skills catalog
|
|
162
157
|
|
|
163
|
-
|
|
158
|
+
CoCo ships **228 skills** (76 core + 152 across bundles)[^installed]. Each one is an instruction set with state, checks, and error handling, not a one-line prompt. A slice:
|
|
164
159
|
|
|
165
|
-
|
|
160
|
+
`ui-ux-pro-max` · `frontend-design` · `brainstorming` · `writing-plans` · `test-driven-development` · `systematic-debugging` · `verification-before-completion` · `prd-generator` · `openai-agents` · `coco-loop` · `coco-ads` · `arch-index`
|
|
166
161
|
|
|
167
|
-
<
|
|
168
|
-
<
|
|
169
|
-
<td width="50%" valign="top">
|
|
162
|
+
<details>
|
|
163
|
+
<summary><strong>▸ Full catalog — every one of the 228 skills</strong></summary>
|
|
170
164
|
|
|
171
|
-
|
|
165
|
+
<br>
|
|
172
166
|
|
|
173
|
-
|
|
167
|
+
**Core skills (76)**
|
|
174
168
|
|
|
175
|
-
|
|
169
|
+
`agent-lightning` · `agent-self-eval` · `ai-marketing-videos` · `ai-product` · `api-design-principles` · `arb-review` · `arch-index` · `axiom-liquid-glass` · `brainstorming` · `browser-automation` · `c4-architecture` · `change-log` · `cli-anything` · `clone-website` · `coco` · `coco-ads` · `coco-cli` · `coco-diagram` · `coco-loop` · `coco-ship` · `code-verification` · `context-budget` · `design-taste-frontend` · `dispatching-parallel-agents` · `doc-sync` · `dr-plan` · `executing-plans` · `expo-api-routes` · `find-skills` · `finishing-a-development-branch` · `frontend-design` · `generate-tests` · `goal` · `humanizer` · `irp` · `journey-map` · `karpathy-guidelines` · `local-llm` · `media-memory` · `meeting-notes` · `nfr-tracker` · `openai-agents` · `openai-api` · `openai-apps-mcp` · `openai-whisper` · `pmstudio` · `prd-generator` · `prd-mastery` · `project-docs` · `receiving-code-review` · `recovery-plan` · `redesign-existing-projects` · `requesting-code-review` · `scroll-world` · `skill-creator` · `skill-evolution` · `stakeholder-comms` · `subagent-driven-development` · `swiftui-liquid-glass` · `systematic-debugging` · `tailwind-patterns` · `task-prd-creator` · `test-driven-development` · `ui-ux-pro-max` · `ultra-think` · `using-git-worktrees` · `using-superpowers` · `vercel-react-best-practices` · `verification-before-completion` · `visual-explainer` · `voice-ai` · `web-design-guidelines` · `workflow-routing` · `writing-plans` · `writing-skills`
|
|
176
170
|
|
|
177
|
-
|
|
178
|
-
<td width="50%" valign="top">
|
|
171
|
+
**GSD bundle skills (68)** — the full `gsd-*` project-orchestration lifecycle: `gsd-new-project`, `gsd-plan-phase`, `gsd-execute-phase`, `gsd-verify-work`, `gsd-autonomous`, `gsd-debug`, `gsd-ui-phase`, `gsd-secure-phase`, `gsd-workstreams`, `gsd-forensics`, `gsd-milestone-summary`, `gsd-map-codebase`, `gsd-profile-user`, and 55 more (see [`systems/gsd/skills/`](systems/gsd/skills/)).
|
|
179
172
|
|
|
180
|
-
|
|
173
|
+
**Brain bundle skills (6)** — `brain` · `brain-init` · `brain-rescan` · `brain-update` · `brain-export` · `brain-wiki`.
|
|
181
174
|
|
|
182
|
-
|
|
175
|
+
**Super Intelligence bundle (13)** — one orchestration skill per built team (ai, engineering, product-design, finance, trading, risk-compliance, strategy, data-analytics, gtm, climate-energy, education-edtech, healthcare-life-sciences, legal-ip), which generate the 342 `/SI-*` commands at install.
|
|
183
176
|
|
|
184
|
-
|
|
177
|
+
**HyperFrames bundle skills (20)** — `embedded-captions` · `faceless-explainer` · `figma` · `general-video` · `hyperframes` · `hyperframes-animation` · `hyperframes-cli` · `hyperframes-core` · `hyperframes-creative` · `hyperframes-keyframes` · `hyperframes-registry` · `media-use` · `motion-graphics` · `music-to-video` · `pr-to-video` · `product-launch-video` · `remotion-to-hyperframes` · `slideshow` · `talking-head-recut` · `website-to-video`.
|
|
185
178
|
|
|
186
|
-
|
|
187
|
-
</tr>
|
|
188
|
-
</table>
|
|
179
|
+
**M0 bundle skills (4)** — `m0` · `m0-handoff` · `m0-recall` · `m0-remember`.
|
|
189
180
|
|
|
190
|
-
**
|
|
181
|
+
**Cognee bundle skills (3)** — `cognee` · `cognee-recall` · `cognee-store`.
|
|
191
182
|
|
|
192
|
-
|
|
183
|
+
**Not installed by default (38):** the `reverse-skill` security pack (33 reverse-engineering and penetration-testing skills, installed only with `--systems reverse-skill`; see [`systems/reverse-skill/`](systems/reverse-skill/)) and 5 Cursor-only skills that ship with the Cursor adapter: `create-rule` · `create-skill` · `create-subagent` · `migrate-to-skills` · `update-cursor-settings`.
|
|
193
184
|
|
|
194
|
-
|
|
195
|
-
/team arch build # map this repository into a committed, commit-pinned index
|
|
196
|
-
/team arch drift # reconcile an existing index against the current tree
|
|
197
|
-
/team arch check # run the validator alone — no model calls, no writes to the index
|
|
198
|
-
/team verify # now includes failure mode (e): architecture abandoned
|
|
199
|
-
/team ship # declares intended architecture, then proves the build matches
|
|
200
|
-
/team ship --no-arch # opt out entirely; every gate reports DISABLED, never a silent pass
|
|
201
|
-
```
|
|
202
|
-
|
|
203
|
-
The approval gate now shows you the intended architecture as a table before the build starts. That is the cheapest moment to disagree with it.
|
|
185
|
+
</details>
|
|
204
186
|
|
|
205
|
-
|
|
187
|
+
## System bundles
|
|
206
188
|
|
|
207
|
-
|
|
189
|
+
Six bundles install by default. `--core-only` skips them. `--systems <list>` installs an explicit subset. `reverse-skill` is not one of the six: it vendors reverse-engineering and penetration-testing methodology under [`systems/reverse-skill/`](systems/reverse-skill/) and installs only with `--systems reverse-skill`. See [`systems/reverse-skill/README.md`](systems/reverse-skill/README.md).
|
|
208
190
|
|
|
209
|
-
|
|
210
|
-
|
|
211
|
-
|
|
212
|
-
|
|
191
|
+
1. **GSD.** An orchestration engine of **68 skills and 24 agents** for phased work on a large codebase. Most GSD skills also need the upstream GSD toolkit at `~/.claude/get-shit-done/`, which this installer does not yet provide; see [`systems/gsd/README.md`](systems/gsd/README.md). Phase state lives in `.planning/`. Workstreams are isolated git checkouts.
|
|
192
|
+
2. **Brain.** A local knowledge-graph engine of **6 skills**. SQLite indexes code entities, decisions, and project terms. The wiki generator writes a page per entity.
|
|
193
|
+
3. **Super Intelligence.** The **495-persona advisory board** and its **342 generated `/SI-*` commands** across 13 teams. This is the board [above](#the-board).
|
|
194
|
+
4. **HyperFrames.** **20 skills** for video and motion, vendored from HeyGen's `hyperframes` under Apache-2.0.
|
|
195
|
+
5. **M0.** **4 skills** for a cross-tool memory thread in one local SQLite file. Python standard library only.
|
|
196
|
+
6. **Cognee.** **3 skills** for a local Cognee graph. They fall back to Brain when `cognee server` is not running.
|
|
213
197
|
|
|
214
|
-
|
|
198
|
+
`/team` is not a bundle. It is part of the core install (`/team ship`, `/team plan`, `/team review`, `/team verify`) and needs no flag. `/team ship` runs 6 development stages and 7 verification gates, and it can declare where files should land, then fail if they did not. The check is structural (paths exist), not a judgement that the code does what the component promised. Design adapted from [lak7/devildev](https://github.com/lak7/devildev) (Apache-2.0); no upstream code was copied. See [`CREDITS.md`](CREDITS.md).
|
|
215
199
|
|
|
216
|
-
|
|
200
|
+
## Technical specifications
|
|
217
201
|
|
|
218
|
-
|
|
202
|
+
A default Claude Code install delivers **190 skills, 387 commands and 34 agents**, measured by running the installer. The repository holds 228 skills; the difference is the opt-in security pack and Cursor-only skills. The measured row is main.
|
|
219
203
|
|
|
220
|
-
|
|
204
|
+
The v1.5.0 release `npx` installs today is older. See [Status and roadmap](#status-and-roadmap).
|
|
221
205
|
|
|
222
206
|
<table align="center">
|
|
223
207
|
<tr>
|
|
224
|
-
<td align="center" width="20%"><h3>
|
|
225
|
-
<td align="center" width="20%"><h3>
|
|
226
|
-
<td align="center" width="20%"><h3>
|
|
208
|
+
<td align="center" width="20%"><h3>228</h3><sub>Skills</sub><br><small>76 Core + 152 Bundle · 190 install by default</small></td>
|
|
209
|
+
<td align="center" width="20%"><h3>387</h3><sub>Slash Commands</sub><br><small>45 Core + 342 Generated</small></td>
|
|
210
|
+
<td align="center" width="20%"><h3>34</h3><sub>Specialized Agents</sub><br><small>10 Core + 24 Bundle</small></td>
|
|
227
211
|
<td align="center" width="20%"><h3>495</h3><sub>Expert Personas</sub><br><small>Super Intelligence Board</small></td>
|
|
228
212
|
<td align="center" width="20%"><h3>15</h3><sub>Cross-IDE Rules</sub><br><small>Cursor MDC Rules</small></td>
|
|
229
213
|
</tr>
|
|
230
214
|
</table>
|
|
231
215
|
|
|
232
216
|
<div align="center">
|
|
233
|
-
<sub><strong>Core install:</strong>
|
|
217
|
+
<sub><strong>Core install:</strong> 146 active assets (76 Skills, 45 Commands, 10 Agents, 15 Rules)</sub><br>
|
|
234
218
|
<sub><strong>A plain install ships every bundle except the security-testing tooling</strong>, which stays opt-in via <code>--systems reverse-skill</code>.</sub><br>
|
|
235
|
-
<sub><strong>Orchestration bundles:</strong> <strong>+68 GSD skills</strong> · <strong>+24 GSD agents</strong> · <strong>+
|
|
219
|
+
<sub><strong>Orchestration bundles:</strong> <strong>+68 GSD skills</strong> · <strong>+24 GSD agents</strong> · <strong>+20 HyperFrames skills</strong> · <strong>+13 Super Intelligence skills</strong> · <strong>+342 SI commands</strong> · <strong>+6 Brain skills</strong> · <strong>+4 M0 skills</strong> · <strong>+3 Cognee skills</strong> · <strong>3 Workflows</strong></sub>
|
|
236
220
|
</div>
|
|
237
221
|
|
|
238
|
-
---
|
|
239
|
-
|
|
240
|
-
<a id="skills-catalog"></a>
|
|
241
|
-
|
|
242
|
-
## Skills Catalog
|
|
243
|
-
|
|
244
|
-
CoCo ships **226 skills** (74 core + 152 across bundles). These are not prompt snippets — each is a full agent instruction set with state management, verification logic, and error handling. A representative slice by category:
|
|
245
|
-
|
|
246
|
-
### Visual design & styling
|
|
247
|
-
`ui-ux-pro-max` (50 styles, 21 palettes, 50 font pairings, 9 stacks) · `frontend-design` · `design-taste-frontend` · `redesign-existing-projects` · `axiom-liquid-glass` (Apple Liquid Glass, WWDC 2025) · `swiftui-liquid-glass` · `web-design-guidelines` · `tailwind-patterns` (Tailwind v4) · `vercel-react-best-practices` · `clone-website` · `c4-architecture` · `arb-review` · `expo-api-routes` · `ai-product` · `scroll-world` (scroll-scrubbed, cut-free camera flythrough landing pages, generated end to end via Higgsfield).
|
|
248
|
-
|
|
249
|
-
### Engineering discipline & quality
|
|
250
|
-
`brainstorming` · `writing-plans` · `executing-plans` · `subagent-driven-development` · `dispatching-parallel-agents` · `test-driven-development` · `systematic-debugging` · `requesting-code-review` · `receiving-code-review` · `verification-before-completion` · `finishing-a-development-branch` · `using-git-worktrees` · `code-verification` (7 bug-vector audit) · `generate-tests` · `api-design-principles` · `cli-anything` · `workflow-routing`.
|
|
251
|
-
|
|
252
|
-
### AI / LLM engineering
|
|
253
|
-
`openai-agents` · `openai-api` · `openai-apps-mcp` · `openai-whisper` · `agent-lightning` (Microsoft Agent Lightning, RL) · `voice-ai` · `ai-marketing-videos`.
|
|
254
|
-
|
|
255
|
-
### Product management & communication
|
|
256
|
-
`prd-generator` · `prd-mastery` · `task-prd-creator` · `project-docs` · `pmstudio` · `nfr-tracker` · `stakeholder-comms` · `meeting-notes` · `change-log`.
|
|
257
|
-
|
|
258
|
-
### Operations, safety & recovery
|
|
259
|
-
`dr-plan` (RTO/RPO) · `irp` (incident response) · `recovery-plan`.
|
|
260
|
-
|
|
261
|
-
### Media, memory & analysis
|
|
262
|
-
`coco-ads` (launch videos via HyperFrames) · `media-memory` (multimodal embeddings) · `doc-sync` · `ultra-think` · `browser-automation`.
|
|
263
|
-
|
|
264
|
-
### The CoCo platform itself
|
|
265
|
-
`coco` (conversational router) · `coco-cli` · `coco-loop` (compiles a plain-language goal into a confirmed charter, then runs a bounded, propose-only autonomous loop) · `skill-creator` · `writing-skills` · `find-skills`.
|
|
266
|
-
|
|
267
|
-
<details>
|
|
268
|
-
<summary><strong>▸ Full catalog — every one of the 226 skills</strong></summary>
|
|
269
|
-
|
|
270
|
-
<br>
|
|
271
|
-
|
|
272
|
-
**Core skills (66)**
|
|
273
|
-
|
|
274
|
-
`agent-lightning` · `ai-marketing-videos` · `ai-product` · `api-design-principles` · `arb-review` · `axiom-liquid-glass` · `brainstorming` · `browser-automation` · `c4-architecture` · `change-log` · `cli-anything` · `clone-website` · `coco` · `coco-ads` · `coco-cli` · `coco-loop` · `code-verification` · `design-taste-frontend` · `dispatching-parallel-agents` · `doc-sync` · `dr-plan` · `executing-plans` · `expo-api-routes` · `find-skills` · `finishing-a-development-branch` · `frontend-design` · `generate-tests` · `irp` · `media-memory` · `meeting-notes` · `nfr-tracker` · `openai-agents` · `openai-api` · `openai-apps-mcp` · `openai-whisper` · `pmstudio` · `prd-generator` · `prd-mastery` · `project-docs` · `receiving-code-review` · `recovery-plan` · `redesign-existing-projects` · `requesting-code-review` · `scroll-world` · `skill-creator` · `stakeholder-comms` · `subagent-driven-development` · `swiftui-liquid-glass` · `systematic-debugging` · `tailwind-patterns` · `task-prd-creator` · `test-driven-development` · `ui-ux-pro-max` · `ultra-think` · `using-git-worktrees` · `using-superpowers` · `vercel-react-best-practices` · `verification-before-completion` · `voice-ai` · `web-design-guidelines` · `workflow-routing` · `writing-plans` · `writing-skills`
|
|
275
|
-
|
|
276
|
-
**GSD bundle skills (68)** — the full `gsd-*` project-orchestration lifecycle: `gsd-new-project`, `gsd-plan-phase`, `gsd-execute-phase`, `gsd-verify-work`, `gsd-autonomous`, `gsd-debug`, `gsd-ui-phase`, `gsd-secure-phase`, `gsd-workstreams`, `gsd-forensics`, `gsd-milestone-summary`, `gsd-map-codebase`, `gsd-profile-user`, and 55 more (see [`systems/gsd/skills/`](systems/gsd/skills/)).
|
|
277
|
-
|
|
278
|
-
**Brain bundle skills (6)** — `brain` · `brain-init` · `brain-rescan` · `brain-update` · `brain-export` · `brain-wiki`.
|
|
279
|
-
|
|
280
|
-
**Super Intelligence bundle (13)** — one orchestration skill per built team (ai, engineering, product-design, finance, trading, risk-compliance, strategy, data-analytics, gtm, climate-energy, education-edtech, healthcare-life-sciences, legal-ip), which generate the 342 `/SI-*` commands at install.
|
|
281
|
-
|
|
282
|
-
</details>
|
|
283
|
-
|
|
284
|
-
---
|
|
285
|
-
|
|
286
|
-
<a id="system-bundles"></a>
|
|
287
|
-
|
|
288
|
-
## System Bundles
|
|
289
|
-
|
|
290
|
-
Bundles are opt-in packages that extend the workspace with specialized databases, pipelines, and rosters. Enable them with `--systems <name>`.
|
|
291
|
-
|
|
292
|
-
One further bundle, `reverse-skill`, is deliberately not part of any default install and is not counted among the four below: it vendors reverse-engineering and penetration-testing methodology under [`systems/reverse-skill/`](systems/reverse-skill/), and it is security tooling that stays installed only when it is asked for by name, with `--systems reverse-skill`. See [`systems/reverse-skill/README.md`](systems/reverse-skill/README.md) for what it contains and how it was curated.
|
|
293
|
-
|
|
294
|
-
### 1. GSD (Get Shit Done)
|
|
295
|
-
An orchestration engine of **68 skills and 24 agents** that manages the lifecycle of complex codebases. A disk-backed phase database records goals, decisions, milestones, and blockers; **workstreams** spin up isolated git checkouts to test refactors without polluting main; **forensics** performs post-mortem root-cause audits when a phase fails; and **autonomous mode** runs parallel waves of subagents through plan milestones end-to-end.
|
|
296
|
-
|
|
297
|
-
### 2. Brain
|
|
298
|
-
A local knowledge-graph engine of **6 skills** connecting email, chat, code, and docs. A local SQLite store indexes code entities, business decisions, project terms, and stakeholder feedback; a **wiki generator** builds hyperlinked articles for every recorded entity; and mail sync links project email threads to related files.
|
|
299
|
-
|
|
300
|
-
### 3. Team
|
|
301
|
-
A multi-agent product-team pipeline (`/team:ship`, `/team:plan`, `/team:review`, `/team:verify`, and more). It spawns specialized subagents — Research, Architect, QA — to execute changes, review diffs, and write tests, with deterministic gates that intercept merges to run lints, import audits, reference checks, and verification suites.
|
|
302
|
-
|
|
303
|
-
### 4. Super Intelligence
|
|
304
|
-
The **495-persona advisory board** and its **342 generated `/SI-*` commands** across 13 teams — the hero capability described [above](#the-board).
|
|
305
|
-
|
|
306
|
-
---
|
|
307
|
-
|
|
308
|
-
## YOLO Mode: True Autonomy
|
|
309
|
-
|
|
310
|
-
<table>
|
|
311
|
-
<tr>
|
|
312
|
-
<td valign="top">
|
|
313
|
-
|
|
314
|
-
```bash
|
|
315
|
-
/coco yolo
|
|
316
|
-
```
|
|
317
|
-
|
|
318
|
-
</td>
|
|
319
|
-
<td valign="top">
|
|
320
|
-
|
|
321
|
-
**Bypass approval gates and let the team run unattended.**<br>
|
|
322
|
-
<sub>Activates autonomous multi-stage execution — pipelines advance phase to phase without manual confirmation. Combine with <code>/gsd-autonomous</code> and <code>--systems gsd</code> to run code migrations or test sweeps overnight. Revert to safety any time with <code>/coco careful</code> or <code>/coco normal</code>.</sub>
|
|
323
|
-
|
|
324
|
-
</td>
|
|
325
|
-
</tr>
|
|
326
|
-
</table>
|
|
327
|
-
|
|
328
|
-
---
|
|
329
|
-
|
|
330
|
-
## Before CoCo vs. After CoCo
|
|
331
|
-
|
|
332
|
-
<table>
|
|
333
|
-
<tr>
|
|
334
|
-
<th width="20%">Scenario</th>
|
|
335
|
-
<th width="38%">Standard AI Assistant (Claude Code / Cursor / Codex)</th>
|
|
336
|
-
<th width="42%">CoCo-Orchestrated Workspace</th>
|
|
337
|
-
</tr>
|
|
338
|
-
<tr>
|
|
339
|
-
<td><strong>Hard decisions</strong></td>
|
|
340
|
-
<td>One model's single opinion, no attribution, no dissent.</td>
|
|
341
|
-
<td>A routed 16–32-person expert panel deliberates and returns an attributed verdict with named dissent.</td>
|
|
342
|
-
</tr>
|
|
343
|
-
<tr>
|
|
344
|
-
<td><strong>Feature development</strong></td>
|
|
345
|
-
<td>Generates a single component, misses side-effects, needs hand-holding.</td>
|
|
346
|
-
<td>Scans repo, designs the interface, plans phases, runs parallel build agents, tests, and verifies.</td>
|
|
347
|
-
</tr>
|
|
348
|
-
<tr>
|
|
349
|
-
<td><strong>Code auditing</strong></td>
|
|
350
|
-
<td>Suggests basic syntax fixes; misses architectural constraints.</td>
|
|
351
|
-
<td>Applies a 7-category audit (TDZ errors, import mismatches, dead code, CSS regressions, mock leaks).</td>
|
|
352
|
-
</tr>
|
|
353
|
-
<tr>
|
|
354
|
-
<td><strong>System debugging</strong></td>
|
|
355
|
-
<td>Repeats similar suggestions; loses track of prior attempts.</td>
|
|
356
|
-
<td>Systematic loop (Reproduce → Isolate → Hypothesize → Fix → Verify) with saved checkpoints.</td>
|
|
357
|
-
</tr>
|
|
358
|
-
<tr>
|
|
359
|
-
<td><strong>Context resets</strong></td>
|
|
360
|
-
<td>Forgets the active plan; starts over from scratch.</td>
|
|
361
|
-
<td>Reads workspace state from disk and resumes the active phase immediately.</td>
|
|
362
|
-
</tr>
|
|
363
|
-
</table>
|
|
364
|
-
|
|
365
|
-
---
|
|
366
|
-
|
|
367
|
-
<a id="see-it-run"></a>
|
|
368
|
-
|
|
369
|
-
## High-Impact Command Demos
|
|
370
|
-
|
|
371
|
-
<table>
|
|
372
|
-
<tr>
|
|
373
|
-
<td width="50%" valign="top">
|
|
374
|
-
|
|
375
|
-
### Cross-team decision
|
|
376
|
-
```bash
|
|
377
|
-
/SI-Decide --debate "Buy vs. build our vector store"
|
|
378
|
-
```
|
|
379
|
-
> Routes a proportional expert panel, deliberates in reacting rounds, and returns an attributed verdict with named dissent.
|
|
380
|
-
|
|
381
|
-
</td>
|
|
382
|
-
<td width="50%" valign="top">
|
|
383
|
-
|
|
384
|
-
### Code-integrity verification
|
|
385
|
-
```bash
|
|
386
|
-
/code-verification
|
|
387
|
-
```
|
|
388
|
-
> Audits fresh code against 7 bug vectors, preventing broken references, imports, and mock leakage.
|
|
389
|
-
|
|
390
|
-
</td>
|
|
391
|
-
</tr>
|
|
392
|
-
<tr>
|
|
393
|
-
<td valign="top">
|
|
394
|
-
|
|
395
|
-
### Website re-engineering
|
|
396
|
-
```bash
|
|
397
|
-
/clone-website https://stripe.com
|
|
398
|
-
```
|
|
399
|
-
> Extracts colors, typography, spacing, shadows, and SVG assets from a live URL into a single-file responsive template.
|
|
400
|
-
|
|
401
|
-
</td>
|
|
402
|
-
<td valign="top">
|
|
403
|
-
|
|
404
|
-
### Systematic debugging
|
|
405
|
-
```bash
|
|
406
|
-
/systematic-debugging
|
|
407
|
-
```
|
|
408
|
-
> Drives a methodical root-cause isolation workflow, preserving hypotheses and findings across API limits.
|
|
409
|
-
|
|
410
|
-
</td>
|
|
411
|
-
</tr>
|
|
412
|
-
<tr>
|
|
413
|
-
<td valign="top">
|
|
414
|
-
|
|
415
|
-
### Spec & PRD generation
|
|
416
|
-
```bash
|
|
417
|
-
/prd-generator
|
|
418
|
-
```
|
|
419
|
-
> Produces a comprehensive PRD — user stories, metrics, risks, architecture.
|
|
420
|
-
|
|
421
|
-
</td>
|
|
422
|
-
<td valign="top">
|
|
423
|
-
|
|
424
|
-
### Launch video
|
|
425
|
-
```bash
|
|
426
|
-
/coco-ads
|
|
427
|
-
```
|
|
428
|
-
> Turns the project you just shipped into a short, shareable launch video, rendered locally via HyperFrames.
|
|
429
|
-
|
|
430
|
-
</td>
|
|
431
|
-
</tr>
|
|
432
|
-
</table>
|
|
433
|
-
|
|
434
|
-
---
|
|
435
|
-
|
|
436
|
-
## Power Features
|
|
437
|
-
|
|
438
|
-
<table>
|
|
439
|
-
<tr>
|
|
440
|
-
<td width="50%" valign="top">
|
|
441
|
-
|
|
442
|
-
### CoCo conversational router
|
|
443
|
-
```bash
|
|
444
|
-
/coco
|
|
445
|
-
```
|
|
446
|
-
> A central dashboard that routes plain-English asks — "what's blocking?", "summarize files", "run tests" — to the right sub-command.
|
|
447
|
-
|
|
448
|
-
</td>
|
|
449
|
-
<td width="50%" valign="top">
|
|
450
|
-
|
|
451
|
-
### Multimodal memory
|
|
452
|
-
```bash
|
|
453
|
-
/media-memory
|
|
454
|
-
```
|
|
455
|
-
> A local embedding database (Gemini Embedding + ChromaDB) that indexes, searches, and associates diagrams, audio, and docs with your code.
|
|
456
|
-
|
|
457
|
-
</td>
|
|
458
|
-
</tr>
|
|
459
|
-
<tr>
|
|
460
|
-
<td valign="top">
|
|
461
|
-
|
|
462
|
-
### Scheduled & loop agents
|
|
463
|
-
```bash
|
|
464
|
-
# Point your host CLI's scheduler at any CoCo command:
|
|
465
|
-
/schedule run /team:review every Monday
|
|
466
|
-
/loop 5m /team:verify
|
|
467
|
-
```
|
|
468
|
-
> CoCo ships no scheduler of its own. Where your host CLI provides scheduling or looping (such as Claude Code's `/schedule` and `/loop`), point it at any CoCo command to run reviews or watches unattended.
|
|
469
|
-
|
|
470
|
-
</td>
|
|
471
|
-
<td valign="top">
|
|
472
|
-
|
|
473
|
-
### Email triage suite
|
|
474
|
-
```bash
|
|
475
|
-
/email:summary
|
|
476
|
-
/email:today
|
|
477
|
-
```
|
|
478
|
-
> Reads your active Outlook instance (Legacy AppleScript or New Outlook MIME cache) to fetch, search, thread, and draft replies to project mail.
|
|
479
|
-
|
|
480
|
-
</td>
|
|
481
|
-
</tr>
|
|
482
|
-
<tr>
|
|
483
|
-
<td valign="top">
|
|
484
|
-
|
|
485
|
-
### Workspace branching
|
|
486
|
-
```bash
|
|
487
|
-
/gsd-workstreams create feature-a
|
|
488
|
-
/gsd-new-workspace
|
|
489
|
-
```
|
|
490
|
-
> Branch development workstreams within one repo — isolated, sandboxed checkouts with separate project state.
|
|
491
|
-
|
|
492
|
-
</td>
|
|
493
|
-
<td valign="top">
|
|
494
|
-
|
|
495
|
-
### AGENTS.md compiler
|
|
496
|
-
```bash
|
|
497
|
-
bash adapters/codex/install.sh
|
|
498
|
-
```
|
|
499
|
-
> Compiles all active rules, skills, and command definitions into a standard `AGENTS.md` recognized by Aider, Cline, and Continue.
|
|
500
|
-
|
|
501
|
-
</td>
|
|
502
|
-
</tr>
|
|
503
|
-
</table>
|
|
504
|
-
|
|
505
|
-
---
|
|
506
|
-
|
|
507
|
-
## How CoCo Works
|
|
508
|
-
|
|
509
|
-
```mermaid
|
|
510
|
-
flowchart LR
|
|
511
|
-
U["You<br/>(Slash command, plain English)"] --> AI["Your AI Harness<br/>(Claude Code, Cursor, Codex…)"]
|
|
512
|
-
AI --> A["IDE Adapter"]
|
|
513
|
-
A --> S["CoCo Assets<br/>skills · agents · commands · rules · personas"]
|
|
514
|
-
S -. Read at runtime .-> AI
|
|
515
|
-
AI --> Agents["Parallel Subagents<br/>(executed by your harness)"]
|
|
516
|
-
Agents --> AI
|
|
517
|
-
```
|
|
518
|
-
|
|
519
|
-
<div align="center"><sub>CoCo is the asset + orchestration layer. Your harness runs the model and executes the agents.</sub></div>
|
|
520
|
-
|
|
521
|
-
---
|
|
522
|
-
|
|
523
|
-
## Installation Matrix
|
|
524
|
-
|
|
525
|
-
<table>
|
|
526
|
-
<tr>
|
|
527
|
-
<td width="50%" valign="top">
|
|
528
|
-
|
|
529
|
-
**Standard 90-second setup**
|
|
530
|
-
|
|
531
|
-
```bash
|
|
532
|
-
git clone https://github.com/coco-research/coco.git
|
|
533
|
-
cd coco
|
|
534
|
-
bash install.sh
|
|
535
|
-
```
|
|
536
|
-
*The installer auto-detects your active IDE/CLI.*
|
|
537
|
-
|
|
538
|
-
</td>
|
|
539
|
-
<td width="50%" valign="top">
|
|
540
|
-
|
|
541
|
-
**Via the npm CLI wrapper**
|
|
542
|
-
|
|
543
|
-
```bash
|
|
544
|
-
# The package name is reserved as `cocosuperintelligence` but is not published
|
|
545
|
-
# on npmjs.com yet (registry 404). Until it is, run the wrapper from a clone:
|
|
546
|
-
git clone --branch v1.5.0 --depth 1 https://github.com/coco-research/coco.git
|
|
547
|
-
node coco/bin/coco.js --help
|
|
548
|
-
```
|
|
549
|
-
*Do not `npx cocosuperintelligence` until the name is published — npx would
|
|
550
|
-
resolve whatever first claims that name. The wrapper clones the pinned
|
|
551
|
-
release tag, not floating `main`.*
|
|
552
|
-
|
|
553
|
-
</td>
|
|
554
|
-
</tr>
|
|
555
|
-
<tr>
|
|
556
|
-
<td valign="top">
|
|
557
|
-
|
|
558
|
-
**Narrow the install (optional)**
|
|
559
|
-
|
|
560
|
-
```bash
|
|
561
|
-
# Every bundle installs by default. To take a subset instead:
|
|
562
|
-
bash install.sh --systems gsd,brain
|
|
563
|
-
bash install.sh --systems superintelligence
|
|
564
|
-
|
|
565
|
-
# Or the core set alone, with no bundles:
|
|
566
|
-
bash install.sh --core-only
|
|
567
|
-
```
|
|
568
|
-
|
|
569
|
-
</td>
|
|
570
|
-
<td valign="top">
|
|
571
|
-
|
|
572
|
-
**Manual adapter overrides**
|
|
573
|
-
|
|
574
|
-
```bash
|
|
575
|
-
bash install.sh --adapter claude-code
|
|
576
|
-
bash install.sh --adapter cursor
|
|
577
|
-
bash install.sh --adapter grok
|
|
578
|
-
bash install.sh --adapter vscode
|
|
579
|
-
bash install.sh --adapter codex
|
|
580
|
-
bash install.sh --adapter pi-desktop
|
|
581
|
-
bash install.sh --adapter generic
|
|
582
|
-
```
|
|
583
|
-
|
|
584
|
-
</td>
|
|
585
|
-
</tr>
|
|
586
|
-
<tr>
|
|
587
|
-
<td colspan="2" valign="top">
|
|
588
|
-
|
|
589
|
-
**One-line remote install**
|
|
590
|
-
|
|
591
|
-
```bash
|
|
592
|
-
bash <(curl -fsSL https://raw.githubusercontent.com/coco-research/coco/main/bin/coco-bootstrap.sh)
|
|
593
|
-
```
|
|
594
|
-
*Clones the pinned release tag (currently `v1.5.0`, not floating `main`) to `~/.coco`, then pauses and prints the commit hash, date, and message so you can verify it against [that release](https://github.com/coco-research/coco/releases/tag/v1.5.0) before it runs `install.sh`. Type `y` to proceed. Pass `--yes` or set `COCO_BOOTSTRAP_YES=1` to skip the prompt for CI or scripted installs, and pass adapter/systems flags after `--`, e.g. `-- --adapter cursor --systems gsd`.*
|
|
595
|
-
|
|
596
|
-
</td>
|
|
597
|
-
</tr>
|
|
598
|
-
</table>
|
|
599
|
-
|
|
600
|
-
Adapter flags, bundle selection, and per-adapter targets: [`docs/install.md`](docs/install.md).
|
|
601
|
-
|
|
602
|
-
The install paths above are deliberately conservative: the remote bootstrap never runs code before you have confirmed the commit it cloned, `install.sh` only ever reads and symlinks files already inside the repository, the CI workflows that build and publish Coco pin their `setup-node` and `setup-python` Actions to commit SHAs rather than mutable tags, and nothing Coco does locally or in CI sends data anywhere. See [`SECURITY.md`](SECURITY.md) for the full policy.
|
|
603
|
-
|
|
604
|
-
---
|
|
605
|
-
|
|
606
|
-
## Staying Current
|
|
607
|
-
|
|
608
|
-
CoCo tells you when a newer version is available — it checks the GitHub repository at most once per day, prints a one-line banner, and sends **no telemetry**. Disable with `COCO_NO_UPDATE_CHECK=1`.
|
|
609
|
-
|
|
610
|
-
<table>
|
|
611
|
-
<tr>
|
|
612
|
-
<td width="50%" valign="top">
|
|
613
|
-
|
|
614
|
-
**git clones**
|
|
615
|
-
|
|
616
|
-
```bash
|
|
617
|
-
# print version + update status
|
|
618
|
-
bash scripts/check-update.sh
|
|
619
|
-
|
|
620
|
-
# apply an update
|
|
621
|
-
git pull --ff-only && bash install.sh
|
|
622
|
-
```
|
|
623
|
-
|
|
624
|
-
</td>
|
|
625
|
-
<td width="50%" valign="top">
|
|
626
|
-
|
|
627
|
-
**CLI wrapper (from a clone)**
|
|
628
|
-
|
|
629
|
-
```bash
|
|
630
|
-
# print version + check for updates
|
|
631
|
-
node bin/coco.js version
|
|
632
|
-
|
|
633
|
-
# apply an update (checks out the pinned release tag)
|
|
634
|
-
node bin/coco.js update
|
|
635
|
-
```
|
|
636
|
-
Until `cocosuperintelligence` is published on npmjs.com, do not use `npx` for this.
|
|
637
|
-
|
|
638
|
-
</td>
|
|
639
|
-
</tr>
|
|
640
|
-
</table>
|
|
641
|
-
|
|
642
|
-
<sub>Releases are tracked in [`CHANGELOG.md`](CHANGELOG.md) and on the <a href="https://github.com/coco-research/coco/releases">GitHub Releases</a> page.</sub>
|
|
643
|
-
|
|
644
|
-
---
|
|
645
|
-
|
|
646
|
-
## System Compatibility
|
|
647
|
-
|
|
648
|
-
<table>
|
|
649
|
-
<tr>
|
|
650
|
-
<th>AI Tool / CLI</th>
|
|
651
|
-
<th>Adapter Name</th>
|
|
652
|
-
<th>Status</th>
|
|
653
|
-
<th>Details</th>
|
|
654
|
-
</tr>
|
|
655
|
-
<tr>
|
|
656
|
-
<td><strong><a href="https://docs.anthropic.com/en/docs/claude-code">Claude Code</a></strong></td>
|
|
657
|
-
<td><code>claude-code</code></td>
|
|
658
|
-
<td>Stable</td>
|
|
659
|
-
<td>Full support for slash commands, agents, and local settings.</td>
|
|
660
|
-
</tr>
|
|
661
|
-
<tr>
|
|
662
|
-
<td><strong><a href="https://cursor.com/">Cursor</a></strong></td>
|
|
663
|
-
<td><code>cursor</code></td>
|
|
664
|
-
<td>Stable</td>
|
|
665
|
-
<td>Links MDC rules and custom workspace skills.</td>
|
|
666
|
-
</tr>
|
|
667
|
-
<tr>
|
|
668
|
-
<td><strong>Grok Build / Grok CLI</strong></td>
|
|
669
|
-
<td><code>grok</code></td>
|
|
670
|
-
<td>Stable</td>
|
|
671
|
-
<td>Links skills, commands, agents and rules into <code>~/.grok/</code>. Registers <code>coco-platform</code> MCP and m0 hooks when those binaries are present.</td>
|
|
672
|
-
</tr>
|
|
673
|
-
<tr>
|
|
674
|
-
<td><strong><a href="https://github.com/openai/codex">Codex CLI</a></strong></td>
|
|
675
|
-
<td><code>codex</code></td>
|
|
676
|
-
<td>Stable</td>
|
|
677
|
-
<td>Compiles assets into root context.</td>
|
|
678
|
-
</tr>
|
|
679
|
-
<tr>
|
|
680
|
-
<td><strong>Aider / Cline / Windsurf</strong></td>
|
|
681
|
-
<td><code>generic</code></td>
|
|
682
|
-
<td>Stable</td>
|
|
683
|
-
<td>Generates a portable, root-level <code>AGENTS.md</code> definition.</td>
|
|
684
|
-
</tr>
|
|
685
|
-
<tr>
|
|
686
|
-
<td><strong><a href="https://code.visualstudio.com/">VS Code</a> / Copilot CLI</strong></td>
|
|
687
|
-
<td><code>vscode</code></td>
|
|
688
|
-
<td>Stable</td>
|
|
689
|
-
<td>Links skills, agents and rules into <code>~/.copilot/</code> and commands into the VS Code prompts folder. No extension needed.</td>
|
|
690
|
-
</tr>
|
|
691
|
-
<tr>
|
|
692
|
-
<td><strong>Antigravity (Google)</strong></td>
|
|
693
|
-
<td><code>antigravity</code></td>
|
|
694
|
-
<td>Planned</td>
|
|
695
|
-
<td>Integration planned for the v0.2 release.</td>
|
|
696
|
-
</tr>
|
|
697
|
-
</table>
|
|
698
|
-
|
|
699
|
-
---
|
|
700
|
-
|
|
701
|
-
## Technical Specifications
|
|
702
|
-
|
|
703
222
|
<table>
|
|
704
223
|
<tr><td><strong>Spec Version</strong></td><td>1.5.0</td></tr>
|
|
705
224
|
<tr><td><strong>License</strong></td><td>Open-core — <a href="LICENSE">MIT</a> core; Super Intelligence is <a href="systems/superintelligence/LICENSE">proprietary</a></td></tr>
|
|
706
|
-
<tr><td><strong>Total Skills</strong></td><td>
|
|
707
|
-
<tr><td><strong>Slash Commands</strong></td><td>
|
|
708
|
-
<tr><td><strong>Specialized Agents</strong></td><td>
|
|
709
|
-
<tr><td><strong>Expert Personas</strong></td><td>495 across 13 departments and
|
|
225
|
+
<tr><td><strong>Total Skills</strong></td><td>228 in the repository (76 Core + 152 Bundle); 190 install by default on Claude Code</td></tr>
|
|
226
|
+
<tr><td><strong>Slash Commands</strong></td><td>387 on a default install — 45 Core (shipped) + 342 Super Intelligence (325 per-team + 17 cross-team, generated at install)</td></tr>
|
|
227
|
+
<tr><td><strong>Specialized Agents</strong></td><td>34 (10 Core + 24 Bundle)</td></tr>
|
|
228
|
+
<tr><td><strong>Expert Personas</strong></td><td>495 across 13 departments and 98 cells</td></tr>
|
|
710
229
|
<tr><td><strong>System Bundles</strong></td><td>6 (GSD, Brain, Cognee, HyperFrames, Super Intelligence, M0) — installed by default. <code>reverse-skill</code>, the security-testing bundle, stays opt-in with <code>--systems reverse-skill</code>. <code>/team</code> is core and needs no flag.</td></tr>
|
|
711
230
|
<tr><td><strong>Cross-IDE Rules</strong></td><td>15 (.mdc files)</td></tr>
|
|
712
231
|
<tr><td><strong>Workflows Defined</strong></td><td>3 (.md pipelines)</td></tr>
|
|
713
|
-
<tr><td><strong>
|
|
714
|
-
<tr><td><strong>
|
|
715
|
-
<tr><td><strong>Telemetry / SaaS</strong></td><td>None — 100% local files</td></tr>
|
|
232
|
+
<tr><td><strong>Install Time</strong></td><td>About 3 minutes on a fresh machine, including the download (measured at 189 seconds).</td></tr>
|
|
233
|
+
<tr><td><strong>Telemetry / SaaS</strong></td><td>No telemetry, no hosted service. One optional version check against GitHub per day (<code>COCO_NO_UPDATE_CHECK=1</code> turns it off)</td></tr>
|
|
716
234
|
</table>
|
|
717
235
|
|
|
718
|
-
<sub>Core install ships
|
|
719
|
-
|
|
720
|
-
---
|
|
236
|
+
<sub>Core install ships 76 skills + 45 commands + 10 agents + 15 rules (146 active assets). A plain <code>bash install.sh</code> delivers 190 skills, 387 commands and 34 agents on Claude Code: every bundle in the default allow-list (GSD, Brain, Cognee, HyperFrames, Super Intelligence, M0) is installed by default. <code>--core-only</code> installs the core set alone, and <code>--systems <list></code> installs an explicit subset — including <code>reverse-skill</code>, the security-testing bundle, which stays opt-in and is never part of the default set. Super Intelligence slash commands are generated locally at install time from the team registries — no command files are transmitted or stored remotely. <code>adapters/INDEX.md</code> records what each adapter actually delivers, measured by running it.</sub>
|
|
721
237
|
|
|
722
|
-
##
|
|
238
|
+
## FAQ
|
|
723
239
|
|
|
724
240
|
<details>
|
|
725
241
|
<summary><strong>Is CoCo an AI model or an agent harness?</strong></summary>
|
|
726
|
-
Neither. CoCo is a portable layer of Markdown and YAML that installs into the agent harness you already use (Claude Code, Cursor, Codex CLI, or any AGENTS.md tool). Your host harness runs the model and executes the agents; CoCo supplies the skills, commands, agents, rules, and persona board that make it act like a coordinated team.
|
|
727
|
-
</details>
|
|
728
242
|
|
|
729
|
-
|
|
730
|
-
<summary><strong>Is my codebase or data sent to a third party?</strong></summary>
|
|
731
|
-
No. CoCo is entirely local — Markdown, configuration, and shell scripts on your machine. Your existing AI tool's privacy policy is unchanged, and CoCo adds no telemetry.
|
|
243
|
+
Neither. It is Markdown and YAML that installs into the harness you already use (Claude Code, Cursor, Codex CLI, or any AGENTS.md tool). The host runs the model and the agents. CoCo supplies the skills, commands, agents, rules, and the persona board.
|
|
732
244
|
</details>
|
|
733
245
|
|
|
734
246
|
<details>
|
|
735
|
-
<summary><strong>
|
|
736
|
-
|
|
247
|
+
<summary><strong>Is my codebase sent anywhere?</strong></summary>
|
|
248
|
+
|
|
249
|
+
CoCo's files stay on your machine. It adds no telemetry and no hosted service. Your editor's own privacy policy is unchanged. The CLI checks GitHub for a newer version at most once a day. `COCO_NO_UPDATE_CHECK=1` turns that off.
|
|
737
250
|
</details>
|
|
738
251
|
|
|
739
252
|
<details>
|
|
740
|
-
<summary><strong>How do I
|
|
741
|
-
|
|
742
|
-
|
|
743
|
-
find ~/.claude ~/.cursor ~/.grok ~/.copilot -type l -lname "${CLONE}/*" -delete</pre>
|
|
744
|
-
This removes only links whose target is inside this clone. A contains-match such as
|
|
745
|
-
<code>*$(pwd)*</code> would also delete links into any path that merely contains the
|
|
746
|
-
folder name (for example <code>coco-research</code> when the clone is <code>coco</code>).
|
|
747
|
-
|
|
748
|
-
For the Hermes adapter, profiles live under <code>~/.hermes/profiles</code>:
|
|
749
|
-
<pre>CLONE="$(pwd)"
|
|
750
|
-
find ~/.hermes/profiles -type l -lname "${CLONE}/*" -delete</pre>
|
|
751
|
-
The appended rules block between <code><!-- coco:rules-start --></code> and
|
|
752
|
-
<code><!-- coco:rules-end --></code> in each profile's CLAUDE.md is stripped automatically on the next install run without the repo.
|
|
253
|
+
<summary><strong>How do I change a skill?</strong></summary>
|
|
254
|
+
|
|
255
|
+
Edit `skills/<name>/SKILL.md`, then run `bash install.sh` so the symlinks pick up the edit.
|
|
753
256
|
</details>
|
|
754
257
|
|
|
755
258
|
<details>
|
|
756
|
-
<summary><strong>What
|
|
757
|
-
|
|
259
|
+
<summary><strong>What if I switch editors?</strong></summary>
|
|
260
|
+
|
|
261
|
+
Run `bash install.sh --adapter cursor` (or the adapter you want). The skills and workflows come with you. The full list is [`adapters/INDEX.md`](adapters/INDEX.md).
|
|
758
262
|
</details>
|
|
759
263
|
|
|
760
|
-
|
|
264
|
+
## Credits
|
|
265
|
+
|
|
266
|
+
CoCo stands on other people's work. The short list:
|
|
761
267
|
|
|
762
|
-
|
|
268
|
+
- **[obra/superpowers](https://github.com/obra/superpowers)** (Jesse Vincent) for the engineering-discipline skills: brainstorming, debugging, TDD, plans, worktrees, review, verification, skill authoring.
|
|
269
|
+
- **[gsd-build/get-shit-done](https://github.com/gsd-build/get-shit-done)** for the 68-skill / 24-agent GSD bundle.
|
|
270
|
+
- **[heygen-com/hyperframes](https://github.com/heygen-com/hyperframes)** (HeyGen, Inc., Apache-2.0) for the 20-skill video bundle under `systems/hyperframes/`.
|
|
271
|
+
- **[JCodesMore/ai-website-cloner-template](https://github.com/JCodesMore/ai-website-cloner-template)** for the structure behind `clone-website`.
|
|
272
|
+
- **[agents.md](https://agents.md/)** for the vendor-neutral context file the adapters follow.
|
|
273
|
+
- **[nickwinder/synthteam](https://github.com/nickwinder/synthteam)** (Nick Winder) for the debate protocol behind `--debate`.
|
|
274
|
+
- **[latent-spaces/brag](https://github.com/latent-spaces/brag)** (Shunit Haviv) and **[HyperFrames](https://www.npmjs.com/package/hyperframes)** (HeyGen) for the launch-video flow behind `coco-ads`.
|
|
275
|
+
- **[leonxlnx/taste-skill](https://github.com/leonxlnx/taste-skill)** (Leon) for `design-taste-frontend` and `redesign-existing-projects`.
|
|
276
|
+
- **[HKUDS/CLI-Anything](https://github.com/HKUDS/CLI-Anything)** for the method behind `cli-anything`.
|
|
277
|
+
- **[Vercel](https://vercel.com/)**: **[vercel-labs/web-interface-guidelines](https://github.com/vercel-labs/web-interface-guidelines)** for `web-design-guidelines`, and **[vercel-labs/agent-skills](https://github.com/vercel-labs/agent-skills)** for `vercel-react-best-practices` (link-only stub).
|
|
763
278
|
|
|
764
|
-
|
|
279
|
+
Full attributions, including audio ([ende.app](https://ende.app), [Kenney.nl](https://kenney.nl)), are in [`CREDITS.md`](CREDITS.md). If something influenced CoCo and is missing, that is a bug: [open an issue](https://github.com/coco-research/coco/issues/new/choose).
|
|
765
280
|
|
|
766
|
-
|
|
767
|
-
- **Good first issues:** [help wanted / good first issue](https://github.com/coco-research/coco/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22) — small, well-scoped tasks for newcomers.
|
|
768
|
-
- **Report a bug / request a feature:** [open an issue](https://github.com/coco-research/coco/issues/new/choose).
|
|
769
|
-
- **Security:** see [`SECURITY.md`](SECURITY.md) for responsible disclosure.
|
|
770
|
-
- **Conduct:** see [`CODE_OF_CONDUCT.md`](CODE_OF_CONDUCT.md).
|
|
281
|
+
## Contributing
|
|
771
282
|
|
|
772
|
-
|
|
283
|
+
CoCo is open-core: the core is MIT-licensed and contributions are welcome. The Super Intelligence System (`systems/superintelligence/`) is proprietary. See [`systems/superintelligence/LICENSE`](systems/superintelligence/LICENSE).
|
|
773
284
|
|
|
774
|
-
|
|
285
|
+
- Start with [`CONTRIBUTING.md`](CONTRIBUTING.md).
|
|
286
|
+
- Good first issues: [help wanted / good first issue](https://github.com/coco-research/coco/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22).
|
|
287
|
+
- Bugs and features: [open an issue](https://github.com/coco-research/coco/issues/new/choose).
|
|
288
|
+
- Security: [`SECURITY.md`](SECURITY.md). Conduct: [`CODE_OF_CONDUCT.md`](CODE_OF_CONDUCT.md).
|
|
775
289
|
|
|
776
|
-
|
|
290
|
+
Releases are in [`CHANGELOG.md`](CHANGELOG.md) and on the [GitHub Releases](https://github.com/coco-research/coco/releases) page.
|
|
777
291
|
|
|
778
|
-
|
|
292
|
+
## Licence
|
|
779
293
|
|
|
780
|
-
-
|
|
781
|
-
- **[gsd-build/get-shit-done](https://github.com/gsd-build/get-shit-done)** — the 68-skill / 24-agent GSD project-orchestration bundle.
|
|
782
|
-
- **[heygen-com/hyperframes](https://github.com/heygen-com/hyperframes)** (HeyGen, Inc., Apache License 2.0) — the 20-skill HyperFrames video/motion bundle vendored under `systems/hyperframes/` (composition engine, CLI, animation, creative direction, registry, and the video-authoring suite).
|
|
783
|
-
- **[JCodesMore/ai-website-cloner-template](https://github.com/JCodesMore/ai-website-cloner-template)** — the site-cloning structure behind `clone-website`.
|
|
784
|
-
- **[agents.md](https://agents.md/)** community — the vendor-neutral agent-context standard CoCo's adapters follow.
|
|
785
|
-
- **[nickwinder/synthteam](https://github.com/nickwinder/synthteam)** (Nick Winder) — the multi-agent debate protocol behind `--debate` deliberation.
|
|
786
|
-
- **[latent-spaces/brag](https://github.com/latent-spaces/brag)** (Shunit Haviv) and **[HyperFrames](https://www.npmjs.com/package/hyperframes)** (HeyGen) — the launch-video flow and local render engine behind `coco-ads`.
|
|
787
|
-
- **[leonxlnx/taste-skill](https://github.com/leonxlnx/taste-skill)** (Leon) — the design-taste enforcement behind `design-taste-frontend` and `redesign-existing-projects`.
|
|
788
|
-
- **[HKUDS/CLI-Anything](https://github.com/HKUDS/CLI-Anything)** — the "wrap any CLI into a structured skill" methodology behind `cli-anything`.
|
|
789
|
-
- **[Vercel](https://vercel.com/)** and **[vercel-labs/web-interface-guidelines](https://github.com/vercel-labs/web-interface-guidelines)** — the source material behind `vercel-react-best-practices` and `web-design-guidelines`.
|
|
294
|
+
Open-core. The core of this repository is MIT, in [`LICENSE`](LICENSE). Super Intelligence (`systems/superintelligence/`) is proprietary, in [`systems/superintelligence/LICENSE`](systems/superintelligence/LICENSE). Third-party bundles keep their own licences: HyperFrames is Apache-2.0, and `reverse-skill` documents its upstream licence in its own README.
|
|
790
295
|
|
|
791
|
-
|
|
296
|
+
[^installed]: 190 of the 228 skills install by default on Claude Code, together with all 387 commands and 34 agents. The other 38 are the opt-in `reverse-skill` security pack (33, installed only with `--systems reverse-skill`) and 5 Cursor-only skills. Measured by running the installer, not by counting files: see [`docs/delivered-counts.json`](docs/delivered-counts.json), and [`adapters/INDEX.md`](adapters/INDEX.md) for every other IDE.
|
package/bin/coco-bootstrap.sh
CHANGED
|
@@ -8,7 +8,9 @@
|
|
|
8
8
|
# COCO_DIR=~/tools/coco bash <(curl -fsSL ...)
|
|
9
9
|
#
|
|
10
10
|
# Flags (passed through to install.sh after --, or set via env):
|
|
11
|
-
# --adapter <name> claude-code | cursor | grok | vscode | codex | generic
|
|
11
|
+
# --adapter <name> claude-code | cursor | grok | vscode | codex | generic |
|
|
12
|
+
# aider | amazon-q | cline | github-copilot-cli | hermes |
|
|
13
|
+
# pi-desktop | roo-code | vscode-continue | windsurf | zed
|
|
12
14
|
# --systems <list> e.g. gsd,brain,team
|
|
13
15
|
# --dry-run preview only, no writes
|
|
14
16
|
# --yes skip the commit-verification prompt (for CI/scripted installs)
|
|
@@ -23,7 +25,7 @@ set -euo pipefail
|
|
|
23
25
|
|
|
24
26
|
REPO_URL="https://github.com/coco-research/coco.git"
|
|
25
27
|
# Keep in sync with package.json version. Clone the last tagged release, not floating main.
|
|
26
|
-
PINNED_TAG="v1.
|
|
28
|
+
PINNED_TAG="v1.6.0" # x-release-please-version
|
|
27
29
|
INSTALL_DIR="${COCO_DIR:-$HOME/.coco}"
|
|
28
30
|
YES="${COCO_BOOTSTRAP_YES:-}"
|
|
29
31
|
PASS_THROUGH=()
|
|
@@ -83,7 +85,13 @@ if [ -d "$INSTALL_DIR/.git" ]; then
|
|
|
83
85
|
git -C "$INSTALL_DIR" checkout --force "$PINNED_TAG"
|
|
84
86
|
fi
|
|
85
87
|
else
|
|
86
|
-
|
|
88
|
+
# bin/coco.js caches its update check in ~/.coco/.update-check.json, so a bare ~/.coco
|
|
89
|
+
# may hold only that regenerable file. Drop it so the clone can land; anything else
|
|
90
|
+
# in the directory is user data and is still refused below.
|
|
91
|
+
if [ -d "$INSTALL_DIR" ] && [ -f "$INSTALL_DIR/.update-check.json" ] && [ "$(ls -A "$INSTALL_DIR")" = ".update-check.json" ]; then
|
|
92
|
+
rm -f "$INSTALL_DIR/.update-check.json"
|
|
93
|
+
fi
|
|
94
|
+
if [ -e "$INSTALL_DIR" ] && ! { [ -d "$INSTALL_DIR" ] && [ -z "$(ls -A "$INSTALL_DIR")" ]; }; then
|
|
87
95
|
echo "Error: $INSTALL_DIR exists but is not a Coco clone." >&2
|
|
88
96
|
echo "Remove it manually or set COCO_DIR to a different path." >&2
|
|
89
97
|
exit 1
|
package/bin/coco.js
CHANGED
|
@@ -103,7 +103,11 @@ Usage:
|
|
|
103
103
|
Update checks contact only github.com (no telemetry); disable with COCO_NO_UPDATE_CHECK=1.
|
|
104
104
|
|
|
105
105
|
Install flags (passed to install.sh):
|
|
106
|
-
--adapter <name> claude-code | cursor | grok | vscode |
|
|
106
|
+
--adapter <name> claude-code | cursor | grok | vscode |
|
|
107
|
+
codex | generic | aider | amazon-q |
|
|
108
|
+
cline | github-copilot-cli | hermes |
|
|
109
|
+
pi-desktop | roo-code | vscode-continue |
|
|
110
|
+
windsurf | zed
|
|
107
111
|
--systems <list> gsd | brain | cognee | hyperframes | m0 |
|
|
108
112
|
superintelligence (comma-separated)
|
|
109
113
|
--dry-run preview, no writes
|
|
@@ -117,8 +121,7 @@ Examples:
|
|
|
117
121
|
|
|
118
122
|
Repo: https://github.com/coco-research/coco
|
|
119
123
|
Install clones the ${PINNED_TAG} release tag, not floating main.
|
|
120
|
-
|
|
121
|
-
git clone or bin/coco-bootstrap.sh over npx.
|
|
124
|
+
npx cocosuperintelligence is the published path (clones the pin, then install).
|
|
122
125
|
`);
|
|
123
126
|
}
|
|
124
127
|
|
|
@@ -180,7 +183,7 @@ function cmdInstall(argv) {
|
|
|
180
183
|
run('bash', [installScript, ...argv]);
|
|
181
184
|
|
|
182
185
|
console.log(`\nDone. Coco installed at ${dir}.`);
|
|
183
|
-
console.log(`Re-
|
|
186
|
+
console.log(`Re-install later: npx cocosuperintelligence install\nUpdate the clone to the pinned release: npx cocosuperintelligence update`);
|
|
184
187
|
}
|
|
185
188
|
|
|
186
189
|
function cmdUpdate(argv) {
|
|
@@ -213,6 +216,9 @@ function cmdUninstall(argv) {
|
|
|
213
216
|
if (fs.existsSync(dir)) {
|
|
214
217
|
fs.rmSync(dir, { recursive: true, force: true });
|
|
215
218
|
}
|
|
219
|
+
// The update notifier's cache is ours; drop it, and ~/.coco if that leaves it empty.
|
|
220
|
+
fs.rmSync(UPDATE_CACHE, { force: true });
|
|
221
|
+
try { fs.rmdirSync(path.dirname(UPDATE_CACHE)); } catch (_) { /* absent or holds other files */ }
|
|
216
222
|
console.log('Uninstalled.');
|
|
217
223
|
}
|
|
218
224
|
|
|
@@ -249,7 +255,12 @@ function main() {
|
|
|
249
255
|
cmdUninstall(rest);
|
|
250
256
|
break;
|
|
251
257
|
default:
|
|
252
|
-
//
|
|
258
|
+
// Flags pass through to install (e.g., npx cocosuperintelligence --adapter cursor).
|
|
259
|
+
// A bare word is an unknown subcommand: reject it before any clone touches disk.
|
|
260
|
+
if (!sub.startsWith('-')) {
|
|
261
|
+
console.error(`Error: unknown command '${sub}'. Run with --help to see the commands.`);
|
|
262
|
+
process.exit(2);
|
|
263
|
+
}
|
|
253
264
|
cmdInstall(argv);
|
|
254
265
|
checkForUpdate(false);
|
|
255
266
|
}
|
package/package.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "cocosuperintelligence",
|
|
3
|
-
"version": "1.
|
|
4
|
-
"description": "CoCo Super Intelligence
|
|
3
|
+
"version": "1.6.0",
|
|
4
|
+
"description": "CoCo Super Intelligence — summon an advisory board of 495 world-class minds inside your AI coding session. A vendor-neutral orchestration layer for Claude Code, Cursor, and Codex: 228 skills (190 installed by default), 387 commands, 34 agents, and disk-persistent state. 100% local, no telemetry. Open-core: MIT core, proprietary Super Intelligence.",
|
|
5
5
|
"bin": {
|
|
6
6
|
"coco": "bin/coco.js"
|
|
7
7
|
},
|