task-pipeline-skill 1.75.0 → 1.76.1
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/CHANGELOG.md +104 -0
- package/README.md +2 -1
- package/SKILL-CARD.md +1 -1
- package/package.json +5 -4
- package/plugins/task-pipeline/.claude-plugin/plugin.json +1 -1
- package/plugins/task-pipeline/commands/project-audit.md +23 -0
- package/plugins/task-pipeline/skills/project-audit/SKILL.md +202 -0
- package/plugins/task-pipeline/skills/project-audit/references/probes.md +199 -0
- package/plugins/task-pipeline/skills/project-audit/scripts/audit.py +1267 -0
- package/plugins/task-pipeline/skills/task-pipeline/references/artifacts.md +8 -0
- package/plugins/task-pipeline/skills/task-pipeline/references/audit.md +8 -0
|
@@ -200,10 +200,18 @@ plugins/task-pipeline/
|
|
|
200
200
|
templates/ # skeletons seeded into a host project
|
|
201
201
|
hygiene.sh docgate.sh # -> scripts/check-hygiene.sh, check-docs.sh
|
|
202
202
|
README.md # the index — and it is the list, not a copy of it
|
|
203
|
+
scripts/graph.py # the work graph, walked so the model never reads it
|
|
204
|
+
skills/evidence-docs/SKILL.md # the ten canons, as a navigator
|
|
205
|
+
skills/project-audit/ # a whole project as the subject
|
|
206
|
+
SKILL.md # the six phases, cold start
|
|
207
|
+
references/probes.md # the probe catalogue, by stack and shape
|
|
208
|
+
scripts/audit.py # discovery, registry, HTML + JSON sidecar
|
|
209
|
+
commands/project-audit.md # /project-audit
|
|
203
210
|
cursor/rules/task-pipeline.mdc # Cursor channel (self-contained rule)
|
|
204
211
|
bin/task-pipeline.js # npx installer (package task-pipeline-skill)
|
|
205
212
|
install.sh # POSIX installer
|
|
206
213
|
test/validate.py # structural validator (npm test)
|
|
214
|
+
test/graph_test.py test/project_audit_test.py # the other two halves of `npm test`
|
|
207
215
|
.github/workflows/{validate,release}.yml # CI + toggleable release automation
|
|
208
216
|
.github/ISSUE_TEMPLATE/ .github/PULL_REQUEST_TEMPLATE.md
|
|
209
217
|
package.json .gitignore
|
|
@@ -34,6 +34,14 @@ whose whole job is "audit X" runs nothing else.
|
|
|
34
34
|
| [`acceptance.md`](acceptance.md) | stage 10 | did everything **on the list** ship, with evidence? |
|
|
35
35
|
| [`loop-guard.md`](loop-guard.md) | any **editing** loop churns | is this pass undoing the last one? |
|
|
36
36
|
| **this file** | any **audit** pass | what is broken or missing that nobody has compared? |
|
|
37
|
+
| the `project-audit` skill | a **whole project** is the subject | what is true of this repository today — from a cold start, with no brief |
|
|
38
|
+
|
|
39
|
+
**`project-audit` is the cold-start caller of this file, not a second copy of
|
|
40
|
+
it.** This ladder needs a REQ row and a module map to walk; most repositories
|
|
41
|
+
have neither, and *"your spine is missing"* is a true first finding and a useless
|
|
42
|
+
audit. That skill discovers what the project is, probes it, reads the production
|
|
43
|
+
evidence a repository cannot hold, and hands phase 4 back here — so the method
|
|
44
|
+
below stays the one place it is written.
|
|
37
45
|
|
|
38
46
|
`loop-guard.md` governs loops that *change* things — the fix loop, a re-entered
|
|
39
47
|
stage. Its trip means a decision is being re-litigated at the wrong altitude. This
|