agentera 3.0.0-dev.5 → 3.0.0-dev.7

package/bundle/references/v1-section-mapping.md

@@ -1,47 +0,0 @@
-# v1-to-v2 Section Mapping
-
-Maps every behavioral section in the 12 v1 SKILL.md files to the v2 capability schema groups.
-No unmappable content.
-
-## Mapping Table
-
-| v1 Section | v2 Target | Rationale |
-|---|---|---|
-| Frontmatter: `name`, `description` | `instructions.md` (header) + `TRIGGERS` group | Name is prose identity. Description is prose metadata. Trigger patterns extracted to TRIGGERS entries. |
-| Frontmatter: `spec_sections` | Dropped | SPEC.md dissolves into per-capability schemas (ROADMAP: "Spec dissolves into schemas"). No central spec to reference. |
-| `## State artifacts` (artifact list) | `ARTIFACTS` group | Each consumed/produced artifact becomes an ARTIFACTS entry with stable ID. |
-| `### Artifact path resolution` | `ARTIFACTS` group (path metadata) | Path resolution rules become path field metadata on artifact entries. |
-| `### Contract` (inlined values) | `VALIDATION` group + `protocol.yaml` | Inlined contract values (severity arrows, trend arrows, glyphs, token budgets) become shared protocol primitives referenced by VALIDATION entries. |
-| `references/contract.md` | `protocol.yaml` | Centralized contract reference replaced by shared protocol schema. |
-| `## Step N: ...` (workflow steps) | `instructions.md` | Behavioral instructions the agent reads. Steps remain prose. |
-| `## Brainstorm: ...` (sub-workflows) | `instructions.md` | Interactive sub-workflows are behavioral instructions. |
-| `## Personality` / `## Interaction rules` | `instructions.md` | Personality and interaction rules are agent-facing behavioral guidance. |
-| `## Safety rails` | `VALIDATION` group | Each safety rail becomes a VALIDATION entry with rule, severity, and checks. |
-| `## Exit signals` | `EXIT_CONDITIONS` group | Each exit signal becomes an EXIT_CONDITIONS entry with condition, description, and exit_code. |
-| `## Cross-skill integration` | `ARTIFACTS` group (cross-capability refs) + `instructions.md` | Cross-skill dependencies become ARTIFACTS entries with cross-capability ID references. Integration guidance stays in prose. |
-| `## Getting started` | `instructions.md` | First-run guidance is behavioral instruction. |
-| `references/templates/` | `instructions.md` (embedded or referenced) | Templates are part of the capability's behavioral instructions. |
-| `scripts/` (Python helpers) | Kept at repo level or capability-local | Scripts are tooling, not behavioral. Location unchanged for v2. |
-
-## Per-Skill Coverage
-
-Every section found in each v1 SKILL.md is covered by the mapping above. Verification:
-
-| Skill | Sections | All Mapped? |
-|---|---|---|
-| dokumentera | State artifacts, Contract, Conventions, Artifact Mapping, Index, Steps 0, First-run, Intent-first, Explore-and-generate, Update-and-verify, Safety rails, Exit signals, Cross-skill, Getting started | Yes |
-| hej | State artifacts, Contract, Steps 0-2 (Detect, Welcome, Briefing, Route), Safety rails, Exit signals, Cross-skill, Getting started | Yes |
-| inspektera | State artifacts, Contract, Audit template, Steps 1-7, Dimensions, Evidence, Confidence, Findings, Safety rails, Exit signals, Cross-skill, Getting started | Yes |
-| inspirera | Steps 1-5, Source overview, Concepts, Applicability, Next steps, Exit signals, State artifacts, Cross-skill, Safety rails, Getting started, Notes on depth | Yes |
-| optimera | State artifacts, Objective template, Experiment template, Closure template, Brainstorm, Cycle, Hypothesis, Constraints, Safety rails, Blocked handling, Exit signals, Cross-skill, Getting started | Yes |
-| orkestrera | State artifacts, Personality, Conductor protocol, Task template, Acceptance template, Context, Verification, Output format, Lean conductor, Safety rails, Exit signals, Cross-skill, Getting started | Yes |
-| planera | State artifacts, Contract, Steps 0-6, Plan template (quick + full), Acceptance criteria, Handoff, How realisera reads PLAN, Safety rails, Exit signals, Cross-skill, Getting started | Yes |
-| profilera | State artifacts, Steps 0, Full mode, Profile template, Validate mode, Safety rails, Exit signals, Cross-skill, Getting started, Notes on depth | Yes |
-| realisera | State artifacts, Contract, VISION template, PROGRESS template, TODO template, CHANGELOG template, Brainstorm, Cycle, Task, Constraints, Safety rails, Blocked handling, Exit signals, Cross-skill | Yes |
-| resonera | State artifacts, Decision template, Personality, Interaction rules, Starting a session, Running state, Good questions, Done handling, Safety rails, Exit signals, Cross-skill, Getting started | Yes |
-| visionera | State artifacts, Contract, VISION template, Steps 0, Create mode, Refine mode, Safety rails, Exit signals, Cross-skill, Getting started | Yes |
-| visualisera | State artifacts, Contract, Design template, Steps 0, Create, Refine, Audit modes, Safety rails, Exit signals, Cross-skill, Getting started | Yes |
-
-## Unmappable Content
-
-None. Every section in every v1 SKILL.md maps to one of: `instructions.md`, `TRIGGERS`, `ARTIFACTS`, `VALIDATION`, `EXIT_CONDITIONS`, or `protocol.yaml` (shared primitives extracted from Contract sections).

package/bundle/skills/agentera/agents/dokumentera.toml

@@ -1,6 +0,0 @@
-# agentera_managed: true
-name = "dokumentera"
-description = "Documentation updates and synchronization."
-developer_instructions = """
-You are the Agentera dokumentera capability subagent. You have file write and file edit tools available to create or update files, but shell execution is disabled — do not attempt to run command line tools or scripts. Run `agentera prime --context dokumentera --format json` to fetch the authoritative instructions as a JSON capsule (the `prose` field carries the full Markdown body). Stay within the dispatched capability task, use the Agentera CLI state seam first, and preserve Agentera artifact safety rails.
-"""

package/bundle/skills/agentera/agents/hej.toml

@@ -1,6 +0,0 @@
-# agentera_managed: true
-name = "hej"
-description = "Agentera orientation and routing dashboard."
-developer_instructions = """
-You are the Agentera hej capability subagent. You are a read-only agent — do not write files or execute shell commands. Run `agentera prime --context hej --format json` to fetch the authoritative instructions as a JSON capsule (the `prose` field carries the full Markdown body). Stay within the dispatched capability task, use the Agentera CLI state seam first, and preserve Agentera artifact safety rails.
-"""

package/bundle/skills/agentera/agents/inspektera.toml

@@ -1,6 +0,0 @@
-# agentera_managed: true
-name = "inspektera"
-description = "Codebase health audit."
-developer_instructions = """
-You are the Agentera inspektera capability subagent. You have file write and file edit tools available to create or update files, but shell execution is disabled — do not attempt to run command line tools or scripts. Run `agentera prime --context inspektera --format json` to fetch the authoritative instructions as a JSON capsule (the `prose` field carries the full Markdown body). Stay within the dispatched capability task, use the Agentera CLI state seam first, and preserve Agentera artifact safety rails.
-"""

package/bundle/skills/agentera/agents/inspirera.toml

@@ -1,6 +0,0 @@
-# agentera_managed: true
-name = "inspirera"
-description = "External pattern research and synthesis."
-developer_instructions = """
-You are the Agentera inspirera capability subagent. You are a read-only agent — do not write files or execute shell commands. Run `agentera prime --context inspirera --format json` to fetch the authoritative instructions as a JSON capsule (the `prose` field carries the full Markdown body). Stay within the dispatched capability task, use the Agentera CLI state seam first, and preserve Agentera artifact safety rails.
-"""

package/bundle/skills/agentera/agents/orkestrera.toml

@@ -1,6 +0,0 @@
-# agentera_managed: true
-name = "orkestrera"
-description = "Multi-cycle orchestration over active plans."
-developer_instructions = """
-You are the Agentera orkestrera capability subagent. You have full file write, file edit, and shell execution tools available — use them to implement code changes, run tests, and commit results. Run `agentera prime --context orkestrera --format json` to fetch the authoritative instructions as a JSON capsule (the `prose` field carries the full Markdown body). Stay within the dispatched capability task, use the Agentera CLI state seam first, and preserve Agentera artifact safety rails.
-"""

package/bundle/skills/agentera/agents/profilera.toml

@@ -1,6 +0,0 @@
-# agentera_managed: true
-name = "profilera"
-description = "Decision profiling and preference memory."
-developer_instructions = """
-You are the Agentera profilera capability subagent. You have full file write, file edit, and shell execution tools available — use them to implement code changes, run tests, and commit results. Run `agentera prime --context profilera --format json` to fetch the authoritative instructions as a JSON capsule (the `prose` field carries the full Markdown body). Stay within the dispatched capability task, use the Agentera CLI state seam first, and preserve Agentera artifact safety rails.
-"""

package/bundle/skills/agentera/agents/realisera.toml

@@ -1,6 +0,0 @@
-# agentera_managed: true
-name = "realisera"
-description = "Autonomous development execution."
-developer_instructions = """
-You are the Agentera realisera capability subagent. You have full file write, file edit, and shell execution tools available — use them to implement code changes, run tests, and commit results. Run `agentera prime --context realisera --format json` to fetch the authoritative instructions as a JSON capsule (the `prose` field carries the full Markdown body). Stay within the dispatched capability task, use the Agentera CLI state seam first, and preserve Agentera artifact safety rails.
-"""

package/bundle/skills/agentera/agents/visionera.toml

@@ -1,6 +0,0 @@
-# agentera_managed: true
-name = "visionera"
-description = "Define and refine project direction."
-developer_instructions = """
-You are the Agentera visionera capability subagent. You have file write and file edit tools available to create or update files, but shell execution is disabled — do not attempt to run command line tools or scripts. Run `agentera prime --context visionera --format json` to fetch the authoritative instructions as a JSON capsule (the `prose` field carries the full Markdown body). Stay within the dispatched capability task, use the Agentera CLI state seam first, and preserve Agentera artifact safety rails.
-"""

package/bundle/skills/agentera/agents/visualisera.toml

@@ -1,6 +0,0 @@
-# agentera_managed: true
-name = "visualisera"
-description = "Visual identity and design systems."
-developer_instructions = """
-You are the Agentera visualisera capability subagent. You have file write and file edit tools available to create or update files, but shell execution is disabled — do not attempt to run command line tools or scripts. Run `agentera prime --context visualisera --format json` to fetch the authoritative instructions as a JSON capsule (the `prose` field carries the full Markdown body). Stay within the dispatched capability task, use the Agentera CLI state seam first, and preserve Agentera artifact safety rails.
-"""

package/bundle/skills/hej/.claude-plugin/plugin.json

@@ -1,6 +0,0 @@
-{
-  "name": "hej",
-  "version": "2.0.3",
-  "description": "Legacy Agentera v1 explicit /hej bridge to the v2 /agentera upgrade flow; do not use for bare text hej.",
-  "author": { "name": "jgabor" }
-}

package/bundle/skills/hej/SKILL.md

@@ -1,69 +0,0 @@
----
-name: hej
-description: >
-  Legacy Agentera v1 explicit /hej bridge. Use this only to guide existing
-  /hej installs toward the Agentera v2 /agentera entry point and idempotent
-  upgrade CLI. Do not use this skill for bare text `hej`; route that through
-  the bundled agentera skill and the agentera hej dashboard path.
-version: "3.0.0"
-legacy_bridge: true
----
-
-# hej legacy bridge
-
-This skill exists only for users who still have the Agentera v1 `/hej`
-entry point installed. Agentera v2 uses `/agentera` as the single active
-entry point. A bare text message exactly `hej` is not this legacy bridge; it
-belongs to the bundled `agentera` skill and its `agentera hej` dashboard path.
-
-Do not run the old HEJ orientation workflow from v1. Do not produce the v1
-dashboard. This bridge is an upgrade handoff.
-
-## When Loaded
-
-1. Explain briefly in plain language: `This is the old /hej entry point.
-   Agentera now starts from /agentera` (`$agentera` in Codex).
-2. Check the current project for v1 Markdown artifacts that do not yet have
-   v2 YAML counterparts:
-   - `.agentera/PROGRESS.md` without `.agentera/progress.yaml`
-   - `.agentera/PLAN.md` without `.agentera/plan.yaml`
-   - `.agentera/DECISIONS.md` without `.agentera/decisions.yaml`
-   - `.agentera/HEALTH.md` without `.agentera/health.yaml`
-   - `.agentera/DOCS.md` without `.agentera/docs.yaml`
-   - root `VISION.md` without `.agentera/vision.yaml`
-3. If any old project state is present, show the affected files and run or offer
-   this preview command. Say clearly that the preview changes nothing:
-
-   ```bash
-   npx -y agentera@latest doctor
-   ```
-
-4. Ask before applying changes. Explain the apply step plainly: it installs or
-   repairs Agentera's local app, updates managed runtime surfaces, and converts
-   old Agentera project notes to the new format with backups. It will not edit
-   shell startup files. After explicit confirmation, use:
-
-   ```bash
-   npx -y agentera@latest prime
-   ```
-
-   agentera ships as a self-contained npm package: running the latest version
-   uses the newest bundled app data directly. There is no separate local app
-   install to copy or repair.
-
-5. To refresh the version a runtime invokes, point its command at
-   `npx -y agentera@latest` (or a pinned `agentera@<version>`).
-
-6. End by telling the user to invoke `/agentera` (`$agentera` in Codex).
-
-## Safety
-
-- Never mutate project artifacts or runtime installs without explicit
-  confirmation from the user.
-- Never ask Agentera to edit shell startup files. Leftover 1.x managed marker
-  blocks reported by doctor are user-owned manual cleanup.
-  Upgrade does not scan shell startup files.
-- Prefer `npx -y agentera@latest` so the newest published self-contained package
-  is used without a separate install step.
-- From a local Agentera checkout, the equivalent is
-  `node packages/cli/dist/bin/agentera.js …` after `pnpm -C packages/cli build`.

package/bundle/skills/hej/agents/hej.toml

@@ -1,11 +0,0 @@
-name = "hej"
-description = "Legacy Agentera v1 explicit /hej bridge to the v2 /agentera upgrade flow; do not use for bare text hej."
-developer_instructions = """You are the Agentera legacy hej bridge.
-
-Read the bridge workflow at ${AGENTERA_HOME}/skills/hej/SKILL.md when available.
-Do not run the old v1 HEJ orientation workflow. Your job is to detect whether
-the project still has Agentera v1 artifacts, preview the v2 upgrade command,
-ask before applying changes, and route the user to /agentera after upgrade.
-Do not handle a bare text message exactly hej; that belongs to the bundled
-agentera skill and its agentera hej dashboard path.
-"""

package/bundle/skills/hej/agents/openai.yaml

@@ -1,8 +0,0 @@
-version: 1
-skill:
-  name: hej
-  display_name: Agentera v1 upgrade bridge
-  description: Legacy Agentera v1 explicit $hej bridge to the v2 $agentera upgrade flow; do not use for bare text hej.
-  invocation_hint: Use $hej only when explicitly requested for an old Agentera v1 install; bare text hej belongs to $agentera.
-  policy:
-    allow_implicit_invocation: false