@dzhechkov/harness-cli 0.3.96 → 0.3.97
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 +17 -1
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -202,6 +202,22 @@ npx @dzhechkov/keysarium init # 7-phase research + comman
|
|
|
202
202
|
| `mcp` | 16 | MCP servers (agentdb, brave-search, gmail, gitlab, comfyui, notion, ...) |
|
|
203
203
|
| `academic` | 5 | Thesis defense (review, questions, doc-check, live defense + answer eval) |
|
|
204
204
|
|
|
205
|
+
#### What the `meta` skills do & how to trigger them
|
|
206
|
+
|
|
207
|
+
Get the whole set with `dz init --target claude-code --preset meta`, or pick one with `dz init --select <skill>`. Skills **auto-activate** on the phrases below (some also expose a `/slash-command`). The headline skills (`design-thinking`, `feature-adr`, `explore`, `knowledge-extractor`) have their own sections; the rest:
|
|
208
|
+
|
|
209
|
+
| Skill | What it's for | How to trigger |
|
|
210
|
+
|-------|---------------|----------------|
|
|
211
|
+
| `audit` | Whole-codebase deep audit → prioritized P0–P3 findings → your approval → scoped fixes | `/audit` · "audit the codebase" / "health check" / "what should we fix next" |
|
|
212
|
+
| `skill-advisor` | Recommends which skills / presets / npx toolkits fit a task (install order + honest gaps) | `/skill-advisor` · "which skill should I use for…" / "какой скилл" |
|
|
213
|
+
| `adversarial-verifier` | False-positive killer: a skeptic refutes a finding and classifies it (TRUE_POSITIVE / FALSE_POSITIVE / …) | "verify this finding" / "is this a real bug" / "false positive check" |
|
|
214
|
+
| `capture-adr` | Records architecture decisions mid-session as minimal MADR-4.0 ADRs (tagged `needs-oversight`) | "record this decision" / "ADR for this" / "capture that as an ADR" |
|
|
215
|
+
| `external-comms-gate` | Screens outbound text (PR / issue / npm / README) for secrets, PII and leaks → SAFE / WARNING / BLOCK | "leak check" / "safe to publish?" / "screen this before publishing" |
|
|
216
|
+
| `context-window-management` | Strategies for context pressure (prune / checkpoint / summarize / delegate); kicks in ~60% capacity | "running out of context" / "manage context" / "compact" |
|
|
217
|
+
| `reflection-loop` | Standalone critique → revise cycle (≤3 rounds) for code, text, architecture or research | `/reflection-loop` · "critique this" / "review and improve" |
|
|
218
|
+
| `structured-reasoning` | Picks the reasoning strategy (Tree-of-Thought / CoT / compression) and checks the conclusion follows | "reason about…" / "explore options" / "compare approaches" |
|
|
219
|
+
| `skill-crystallizer` | Auto-creates skills from execution traces, combines skills, and repairs broken ones | "create skill from this" / "combine skills" / "fix skill" |
|
|
220
|
+
|
|
205
221
|
### Standalone Packages (install via npx, no dz CLI needed)
|
|
206
222
|
|
|
207
223
|
| Package | Install | What it does |
|
|
@@ -956,7 +972,7 @@ npx @dzhechkov/p-replicator init
|
|
|
956
972
|
|
|
957
973
|
## Status
|
|
958
974
|
|
|
959
|
-
`v0.3.
|
|
975
|
+
`v0.3.96` — published on npm. Also available as [Claude Plugin](#claude-plugin). Part of [DZ Harness Hub](https://github.com/djd1m/dz-harness-hub).
|
|
960
976
|
|
|
961
977
|
## Claude Plugin
|
|
962
978
|
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@dzhechkov/harness-cli",
|
|
3
|
-
"version": "0.3.
|
|
3
|
+
"version": "0.3.97",
|
|
4
4
|
"description": "The dz CLI — install AI skills for Claude Code, Codex, OpenCode, Hermes, OpenClaude, GitHub Copilot. 32 commands, 11 presets, 6 platform targets.",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"license": "MIT",
|