xtrm-tools 0.7.12 → 0.7.13
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/.xtrm/hooks/specialists/specialists-memory-cache-sync.mjs +57 -0
- package/.xtrm/registry.json +477 -389
- package/.xtrm/skills/default/premortem/SKILL.md +218 -0
- package/.xtrm/skills/default/releasing/SKILL.md +90 -0
- package/.xtrm/skills/default/sync-docs/SKILL.md +88 -208
- package/.xtrm/skills/default/sync-docs/scripts/pre-context.sh +17 -0
- package/.xtrm/skills/default/update-specialists/SKILL.md +228 -36
- package/.xtrm/skills/default/update-xt/SKILL.md +34 -0
- package/.xtrm/skills/default/using-kpi/SKILL.md +150 -0
- package/.xtrm/skills/default/using-specialists-v2/SKILL.md +683 -0
- package/cli/dist/index.cjs +839 -429
- package/cli/dist/index.cjs.map +1 -1
- package/cli/package.json +1 -1
- package/package.json +2 -2
- package/packages/pi-extensions/.serena/project.yml +119 -0
- package/packages/pi-extensions/extensions/pi-serena-compact/index.ts +4 -12
- package/packages/pi-extensions/extensions/xtrm-loader/index.ts +0 -1
- package/packages/pi-extensions/extensions/xtrm-ui/index.ts +201 -36
- package/packages/pi-extensions/extensions/xtrm-ui/themes/pidex-dark-flattools.json +79 -0
- package/packages/pi-extensions/extensions/xtrm-ui/themes/pidex-dark.json +85 -0
- package/packages/pi-extensions/extensions/xtrm-ui/themes/pidex-light-flattools.json +79 -0
- package/packages/pi-extensions/extensions/xtrm-ui/themes/pidex-light.json +85 -0
- package/packages/pi-extensions/package.json +1 -1
- package/packages/pi-extensions/themes/xtrm-ui/pidex-dark-flattools.json +79 -0
- package/packages/pi-extensions/themes/xtrm-ui/pidex-dark.json +3 -3
- package/packages/pi-extensions/themes/xtrm-ui/pidex-light-flattools.json +79 -0
|
@@ -6,35 +6,56 @@ description: >
|
|
|
6
6
|
"sp is out of date", "hooks not firing", "skills not loading after update",
|
|
7
7
|
or when drift is detected in installed specialists config, hooks, jobs, DB,
|
|
8
8
|
extensions, or worktree cleanup.
|
|
9
|
-
version: 1.
|
|
10
|
-
synced_at:
|
|
9
|
+
version: 1.4
|
|
10
|
+
synced_at: 2026-04-29
|
|
11
11
|
---
|
|
12
12
|
|
|
13
13
|
# update-specialists
|
|
14
14
|
|
|
15
15
|
Bring specialists install back to canonical state. Detect drift, apply targeted
|
|
16
|
-
fixes, then verify with `sp doctor`.
|
|
16
|
+
fixes, then verify with `sp doctor`. Treat canonical state as both:
|
|
17
|
+
1. healthy repo wiring and runtime behavior, and
|
|
18
|
+
2. parity with currently installed `@jaggerxtrm/specialists` package version
|
|
19
|
+
when package-level comparison is available.
|
|
20
|
+
|
|
21
|
+
Ownership contract during repair:
|
|
22
|
+
- upstream source: package `config/*` (read-only for repo operators)
|
|
23
|
+
- managed mirror: `.specialists/default/*` (refresh via `sp init --sync-defaults`; sync scope = specialists + mandatory-rules + nodes; no hand edits)
|
|
24
|
+
- repo custom layer: `.specialists/user/*` + `config/nodes/*` + `.specialists/mandatory-rules/*` (rule overlay, wins on set-id conflict; NOT drift — do not overwrite or flag)
|
|
25
|
+
- runtime/generated: `.specialists/{jobs,ready,db}`
|
|
26
|
+
|
|
27
|
+
Isolation rule: backlog-clean surfaces out of scope for this skill.
|
|
17
28
|
|
|
18
29
|
## Canonical State
|
|
19
30
|
|
|
20
31
|
Check each item explicitly. This is what a healthy specialists-initialized project
|
|
21
32
|
looks like.
|
|
22
33
|
|
|
34
|
+
### Package + runtime parity
|
|
35
|
+
|
|
36
|
+
| Check | Expected value |
|
|
37
|
+
|-------|----------------|
|
|
38
|
+
| Installed `@jaggerxtrm/specialists` package version | Matches intended runtime version for repo install |
|
|
39
|
+
| `sp --version` / `specialists --version` | Matches installed package version or same release line |
|
|
40
|
+
| Installed package root | Resolvable from Node / npm environment |
|
|
41
|
+
| Canonical package defaults | Available from installed package for direct diffing |
|
|
42
|
+
| Repo install vs package install | No unexpected drift in canonical files unless intentionally customized |
|
|
43
|
+
|
|
23
44
|
### Specialists configs
|
|
24
45
|
|
|
25
46
|
| Check | Expected value |
|
|
26
47
|
|-------|----------------|
|
|
27
48
|
| `.specialists/default/*.specialist.json` | JSON-first specialist configs present |
|
|
28
49
|
| `metadata.name` | Matches filename stem |
|
|
29
|
-
| `metadata.version` | Valid semver string |
|
|
50
|
+
| `metadata.version` | Valid semver string and consistent with canonical shipped copy when comparing like-for-like |
|
|
30
51
|
| `metadata.description` | Present |
|
|
31
52
|
| `metadata.category` | Present |
|
|
32
53
|
| `execution.model` | Present and pingable |
|
|
33
54
|
| `execution.fallback_model` | Present, different provider from primary |
|
|
34
55
|
| `execution.permission_required` | Valid enum |
|
|
35
|
-
| `
|
|
36
|
-
| `execution.extensions.gitnexus` | Present when skill needs opt-out or default true |
|
|
56
|
+
| `skills.paths` | Referenced skill paths resolve correctly |
|
|
37
57
|
| `execution.interactive` | Matches intended keep-alive behavior |
|
|
58
|
+
| Installed default specialist copy | Matches canonical package copy unless intentionally customized |
|
|
38
59
|
|
|
39
60
|
### Hooks wiring
|
|
40
61
|
|
|
@@ -44,6 +65,7 @@ looks like.
|
|
|
44
65
|
| Hook events | At minimum: `SessionStart`, `PreToolUse`, `PostToolUse`, `Stop` |
|
|
45
66
|
| Hook paths | Point at specialists runtime hook scripts, not stale xtrm-only paths |
|
|
46
67
|
| Hook format | Matches project's installed settings format and loads cleanly |
|
|
68
|
+
| Installed hook scripts | Match canonical package hook files unless intentionally customized |
|
|
47
69
|
|
|
48
70
|
### CLI reachability
|
|
49
71
|
|
|
@@ -68,20 +90,69 @@ looks like.
|
|
|
68
90
|
|
|
69
91
|
| Check | Expected value |
|
|
70
92
|
|-------|----------------|
|
|
71
|
-
| specialists DB | Opens cleanly |
|
|
72
|
-
| Schema version | Matches runtime expectation |
|
|
93
|
+
| specialists DB | Opens cleanly (`.specialists/db/observability.db`) |
|
|
94
|
+
| Schema version | Matches runtime expectation (current: v11; auto-migrates on next runtime startup) |
|
|
95
|
+
| `specialist_job_metrics` table | Present at v11+ — holds aggregated per-job metrics |
|
|
96
|
+
| `specialist_job_metrics` columns | Includes `active_runtime_ms` + `waiting_ms` (drs41.1 — auto-added by idempotent `migrateToV11` ALTER TABLE on first start of upgraded runtime; pre-existing rows get NULL until next aggregate) |
|
|
97
|
+
| Auto-aggregation hook | Supervisor + `sp stop` invoke `aggregateJobMetricsBestEffort` after terminal-status persistence (drs41.1) — table populates without manual `sp db extract` under normal operation |
|
|
98
|
+
| Merge target lookup | DB-first (post-ofjvj): `readAllJobStatuses()` reads `specialist_jobs` via `listStatuses()`. `sp merge` no longer reads `.specialists/jobs/<id>/status.json`. Older versions silently failed after `sp stop` cleaned status.json. |
|
|
73
99
|
| WAL / busy timeout settings | Present when runtime uses SQLite |
|
|
74
100
|
| Corruption / lock errors | None in `sp doctor` |
|
|
101
|
+
| Pre-prune extract | `sp db prune --apply` extracts metrics to `specialist_job_metrics` before deleting events |
|
|
102
|
+
| Extract backfill | `sp db extract --all-missing` populates metrics for jobs whose events still exist (still useful for backfilling historical jobs that ran before the auto-aggregate hook landed) |
|
|
103
|
+
| Historical stats query | `sp db stats [--spec <name>] [--model <glob>] [--since <dur>]` reads the aggregated table; output includes `active_s`, `waiting_s`, `total_s` (drs41.1) |
|
|
75
104
|
|
|
76
|
-
|
|
105
|
+
**Safety: `sp init` and `sp init --sync-defaults` do NOT touch `.specialists/db/observability.db`.** Init checks file existence and skips with "observability database already exists (not touched)" when present. Schema migrations run on next runtime startup (any `sp` invocation that opens the DB), additively via `ALTER TABLE ADD COLUMN`. No data loss path during a normal package upgrade.
|
|
106
|
+
|
|
107
|
+
### Skills + extensions parity
|
|
77
108
|
|
|
78
109
|
| Check | Expected value |
|
|
79
110
|
|-------|----------------|
|
|
111
|
+
| `.xtrm/skills/default/` | Matches canonical package skill set for installed version |
|
|
112
|
+
| Active skill links / copies | Resolve to expected default or active targets |
|
|
113
|
+
| Skill frontmatter `version` / `synced_at` | Present and reasonable for shipped skills |
|
|
80
114
|
| `quality-gates` | Registered if project uses quality gates |
|
|
81
115
|
| `pi-gitnexus` | Registered when GitNexus integration is expected |
|
|
82
116
|
| `pi-serena-tools` | Registered when Serena integration is expected |
|
|
83
117
|
| Extension paths | Resolve from installed project, not stale workspace copies |
|
|
84
118
|
|
|
119
|
+
### Mandatory-rules template parity (three-tier)
|
|
120
|
+
|
|
121
|
+
Loader unions indexes from three paths and probes set files in reverse precedence
|
|
122
|
+
(overlay wins on set-id conflict). Full authoring guide:
|
|
123
|
+
`config/mandatory-rules/README.md`.
|
|
124
|
+
|
|
125
|
+
| Check | Expected value |
|
|
126
|
+
|-------|----------------|
|
|
127
|
+
| `.specialists/default/mandatory-rules/*` | Mirrors canonical package templates after `sp init --sync-defaults` (managed mirror, no hand edits) |
|
|
128
|
+
| `.specialists/mandatory-rules/*` | Repo-specific overlay (user-maintained). Present when repo ships its own rules. NOT drift. |
|
|
129
|
+
| Template frontmatter | YAML frontmatter present and parseable |
|
|
130
|
+
| `specialist.mandatory_rules.template_sets` references | Resolve in order: `.specialists/mandatory-rules/` → `.specialists/default/mandatory-rules/` → `config/mandatory-rules/` |
|
|
131
|
+
| Index files (`index.json`) | Any of the three tiers may define `required_template_sets` / `default_template_sets`; loader unions + dedups |
|
|
132
|
+
| Prompt injection behavior | Runner appends resolved `MANDATORY_RULES` block at end of prompt; supervisor emits `mandatory_rules_injection` meta event |
|
|
133
|
+
|
|
134
|
+
## Discover Latest Release
|
|
135
|
+
|
|
136
|
+
Before reconciling, determine whether a newer release is published. Compare local `package.json` version to the most recent `vX.Y.Z` tag on `origin`:
|
|
137
|
+
|
|
138
|
+
```bash
|
|
139
|
+
LOCAL=$(node -p "require('./package.json').version")
|
|
140
|
+
LATEST=$(git ls-remote --tags --refs origin | grep -oE 'v[0-9]+\.[0-9]+\.[0-9]+$' | sort -V | tail -1 | sed 's/^v//')
|
|
141
|
+
echo "local: $LOCAL latest: $LATEST"
|
|
142
|
+
```
|
|
143
|
+
|
|
144
|
+
If `LATEST > LOCAL`, read the corresponding `CHANGELOG.md` section to summarize what shipped:
|
|
145
|
+
|
|
146
|
+
```bash
|
|
147
|
+
awk -v ver="$LATEST" '/^## \[v?'"$LATEST"'\]/,/^## \[/{print}' CHANGELOG.md | head -60
|
|
148
|
+
```
|
|
149
|
+
|
|
150
|
+
Surface a one-line summary to the user (Added/Changed/Fixed counts plus the headline) and **ask before pulling**. The reconcile flow below applies regardless of whether the user pulls a new release first or stays on the current version — drift detection is independent of release version.
|
|
151
|
+
|
|
152
|
+
Skip this discovery step entirely when `SPECIALISTS_OFFLINE=1` is set, when offline, or when the user already specified the version. The `using-specialists-v2` skill performs the same lightweight check on session-load and may have already surfaced the notice; do not repeat it.
|
|
153
|
+
|
|
154
|
+
After the user confirms a pull (e.g. `git fetch && git pull origin master`), proceed with detection below to catch any drift introduced by the new release.
|
|
155
|
+
|
|
85
156
|
## Detection
|
|
86
157
|
|
|
87
158
|
Run these in order. Report which checks pass and which drift.
|
|
@@ -93,26 +164,61 @@ sp doctor
|
|
|
93
164
|
# 2. Runtime status
|
|
94
165
|
sp status
|
|
95
166
|
|
|
96
|
-
# 3.
|
|
167
|
+
# 3. Installed package + CLI version parity
|
|
168
|
+
npm ls @jaggerxtrm/specialists --depth=0 2>/dev/null || true
|
|
169
|
+
node -e "try { const pkg=require(require.resolve('@jaggerxtrm/specialists/package.json')); console.log(JSON.stringify({installed_package_version: pkg.version}, null, 2)); } catch (err) { console.log('PACKAGE_NOT_RESOLVABLE'); }"
|
|
170
|
+
sp --version 2>/dev/null || true
|
|
171
|
+
specialists --version 2>/dev/null || true
|
|
172
|
+
|
|
173
|
+
# 4. Resolve canonical package root for direct drift diff
|
|
174
|
+
node -e "try { const path=require('path'); const pkgPath=require.resolve('@jaggerxtrm/specialists/package.json'); console.log(path.dirname(pkgPath)); } catch (err) { console.log('PACKAGE_ROOT_UNAVAILABLE'); }"
|
|
175
|
+
|
|
176
|
+
# 5. Config shape
|
|
97
177
|
find .specialists/default -maxdepth 1 -name '*.specialist.json' -print
|
|
98
178
|
|
|
99
|
-
#
|
|
100
|
-
node -e "const fs=require('fs'); const path=require('path'); const dir='.specialists/default'; for (const file of fs.readdirSync(dir)) { if (!file.endsWith('.specialist.json')) continue; const
|
|
179
|
+
# 6. Validate specialist JSON files
|
|
180
|
+
node -e "const fs=require('fs'); const path=require('path'); const dir='.specialists/default'; for (const file of fs.readdirSync(dir)) { if (!file.endsWith('.specialist.json')) continue; const s=JSON.parse(fs.readFileSync(path.join(dir,file),'utf8')); const m=s.metadata||{}; const e=s.execution||{}; const missing=[]; for (const key of ['name','version','description','category']) if (!m[key]) missing.push('metadata.'+key); for (const key of ['model','fallback_model','permission_required']) if (!e[key]) missing.push('execution.'+key); if (missing.length) console.log(file+': MISSING '+missing.join(', ')); if (m.name && m.name !== file.replace(/\.specialist\.json$/, '')) console.log(file+': NAME MISMATCH '+m.name); }"
|
|
181
|
+
|
|
182
|
+
# 7. Validate referenced skill paths
|
|
183
|
+
node -e "const fs=require('fs'); const path=require('path'); const dir='.specialists/default'; for (const file of fs.readdirSync(dir)) { if (!file.endsWith('.specialist.json')) continue; const s=JSON.parse(fs.readFileSync(path.join(dir,file),'utf8')); for (const p of (s.skills?.paths ?? [])) { if (!fs.existsSync(p)) console.log(file+': MISSING SKILL PATH '+p); } }"
|
|
184
|
+
|
|
185
|
+
# 8. Compare repo defaults against installed package defaults (if package root resolvable)
|
|
186
|
+
PKG_ROOT="$(node -e "try { const path=require('path'); process.stdout.write(path.dirname(require.resolve('@jaggerxtrm/specialists/package.json'))); } catch (err) {}")"
|
|
187
|
+
if [ -n "$PKG_ROOT" ]; then
|
|
188
|
+
diff -rq .specialists/default "$PKG_ROOT/config/specialists" || true
|
|
189
|
+
diff -rq .xtrm/skills/default "$PKG_ROOT/config/skills" || true
|
|
190
|
+
diff -rq .claude/hooks "$PKG_ROOT/config/hooks" || true
|
|
191
|
+
else
|
|
192
|
+
echo PACKAGE_COMPARE_UNAVAILABLE
|
|
193
|
+
fi
|
|
101
194
|
|
|
102
|
-
#
|
|
195
|
+
# 9. Hooks wiring
|
|
103
196
|
node -e "const fs=require('fs'); const p='.claude/settings.json'; if (fs.existsSync(p)) { const s=JSON.parse(fs.readFileSync(p,'utf8')); console.log(JSON.stringify(s.hooks ?? s, null, 2)); } else { console.log('MISSING .claude/settings.json'); }"
|
|
104
197
|
|
|
105
|
-
#
|
|
198
|
+
# 10. Command availability
|
|
106
199
|
command -v sp
|
|
107
200
|
command -v specialists
|
|
201
|
+
specialists init --help | sed -n '1,120p'
|
|
202
|
+
specialists edit --help | sed -n '1,120p' | grep -E -- '--fork-from|fork-from' || true
|
|
108
203
|
sp doctor --json 2>/dev/null || true
|
|
109
204
|
|
|
110
|
-
#
|
|
205
|
+
# 11. Jobs and worktrees
|
|
111
206
|
ls -1 .specialists/jobs 2>/dev/null || true
|
|
112
207
|
find .worktrees -maxdepth 2 -mindepth 1 -type d 2>/dev/null || true
|
|
113
208
|
|
|
114
|
-
#
|
|
209
|
+
# 12. Extension registration
|
|
115
210
|
node -e "const fs=require('fs'); const p='.pi/settings.json'; if (fs.existsSync(p)) console.log(JSON.stringify(JSON.parse(fs.readFileSync(p,'utf8')).skills ?? JSON.parse(fs.readFileSync(p,'utf8')).extensions ?? {}, null, 2)); else console.log('MISSING .pi/settings.json')"
|
|
211
|
+
|
|
212
|
+
# 13a. Observability schema + metrics coverage + drs41.1 column presence
|
|
213
|
+
node -e "const {Database} = require('bun:sqlite'); const p='.specialists/db/observability.db'; const fs=require('fs'); if (!fs.existsSync(p)) { console.log('NO_DB'); process.exit(0); } const db=new Database(p,{readonly:true}); const v=db.query(\"SELECT value FROM schema_meta WHERE key='version'\").get(); const has=db.query(\"SELECT name FROM sqlite_master WHERE type='table' AND name='specialist_job_metrics'\").get(); const jobs=db.query('SELECT COUNT(*) c FROM specialist_jobs').get(); const metrics=has ? db.query('SELECT COUNT(*) c FROM specialist_job_metrics').get() : null; const cols=has ? new Set(db.query('PRAGMA table_info(specialist_job_metrics)').all().map(r=>r.name)) : new Set(); const drs41Cols={active_runtime_ms: cols.has('active_runtime_ms'), waiting_ms: cols.has('waiting_ms')}; console.log(JSON.stringify({schema_version: v?.value, has_metrics_table: !!has, drs41_columns_present: drs41Cols, jobs: jobs.c, metrics_rows: metrics?.c ?? 0, metrics_coverage: metrics ? (metrics.c/jobs.c).toFixed(2) : null}, null, 2));" 2>/dev/null || echo "REQUIRES_BUN_RUNTIME"
|
|
214
|
+
|
|
215
|
+
# 13. Mandatory-rules template tiers + reference checks (three-tier resolution)
|
|
216
|
+
find .specialists/default/mandatory-rules -maxdepth 1 -type f 2>/dev/null || true
|
|
217
|
+
find .specialists/mandatory-rules -maxdepth 1 -type f 2>/dev/null || true
|
|
218
|
+
node -e "const fs=require('fs'); const path=require('path'); const roots=['.specialists/default/specialists','.specialists/user/specialists']; const missing=[]; for (const root of roots) { if (!fs.existsSync(root)) continue; for (const file of fs.readdirSync(root)) { if (!file.endsWith('.specialist.json')) continue; const spec=JSON.parse(fs.readFileSync(path.join(root,file),'utf8')); const sets=spec.specialist?.mandatory_rules?.template_sets ?? []; for (const set of sets) { const candidates=[path.join('.specialists/mandatory-rules',set+'.md'), path.join('.specialists/default/mandatory-rules',set+'.md'), path.join('config/mandatory-rules',set+'.md')]; if (!candidates.some((p)=>fs.existsSync(p))) missing.push(file+': missing template set '+set); } } } if (missing.length) console.log(missing.join('\n'));"
|
|
219
|
+
|
|
220
|
+
# 14. Shipped skill frontmatter parity
|
|
221
|
+
node -e "const fs=require('fs'); const path=require('path'); const dir='.xtrm/skills/default'; if (!fs.existsSync(dir)) process.exit(0); for (const name of fs.readdirSync(dir)) { const p=path.join(dir,name,'SKILL.md'); if (!fs.existsSync(p)) continue; const head=fs.readFileSync(p,'utf8').split('---')[1] || ''; const version=(head.match(/version:\s*([^\n]+)/)||[])[1]; const synced=(head.match(/synced_at:\s*([^\n]+)/)||[])[1]; console.log(name+': version='+(version||'missing')+' synced_at='+(synced||'missing')); }"
|
|
116
222
|
```
|
|
117
223
|
|
|
118
224
|
## Drift -> Fix Mapping
|
|
@@ -121,37 +227,75 @@ Use targeted fixes first. Escalate to full sync only if needed.
|
|
|
121
227
|
|
|
122
228
|
| Drift | Fix |
|
|
123
229
|
|-------|-----|
|
|
124
|
-
|
|
|
125
|
-
|
|
|
126
|
-
|
|
|
230
|
+
| Installed package version mismatch | reinstall / upgrade `@jaggerxtrm/specialists`, then re-run checks |
|
|
231
|
+
| CLI version mismatch vs package | reinstall runtime so `sp` / `specialists` align with installed package |
|
|
232
|
+
| Specialist JSON missing required fields | `sp edit <name> ...` or regenerate via `specialists init --sync-defaults` |
|
|
233
|
+
| Need user-layer override from default/package specialist | `sp edit <name> --fork-from <base>` to materialize editable copy in `.specialists/user/` |
|
|
234
|
+
| Specialist JSON schema mismatch | `specialists init --sync-defaults` (refreshes specialists + mandatory-rules + nodes) |
|
|
235
|
+
| Installed specialist default differs from canonical package copy | `specialists init --sync-defaults` unless local customization is intentional |
|
|
236
|
+
| Hooks missing or stale | `specialists init` |
|
|
237
|
+
| Installed hook file differs from canonical package copy | `specialists init` unless local customization is intentional |
|
|
127
238
|
| `sp` / `specialists` missing from PATH | Reinstall / re-bootstrap specialists runtime |
|
|
128
|
-
| Job dir missing | `
|
|
239
|
+
| Job dir missing | `specialists init` |
|
|
129
240
|
| Orphaned `.worktrees/` entries | `specialists clean` |
|
|
130
|
-
| SQLite schema/version mismatch | `sp doctor` first, then `
|
|
131
|
-
|
|
|
132
|
-
|
|
|
241
|
+
| SQLite schema/version mismatch | `sp doctor` first, then `specialists init --sync-defaults` or runtime migration command |
|
|
242
|
+
| Schema below v11 (no `specialist_job_metrics`) | Reinstall / upgrade runtime; table is created by initSchema / migrateToV11. No data loss — raw events untouched. |
|
|
243
|
+
| `specialist_job_metrics` missing `active_runtime_ms` / `waiting_ms` columns (post-drs41.1) | Open any `sp` command — `migrateToV11` is idempotent and ALTERs the table to add the columns. No reinstall needed. Pre-existing rows show NULL until next aggregate or `sp db extract --all-missing`. |
|
|
244
|
+
| Auto-aggregate hook absent (older runtime) — empty `specialist_job_metrics` despite job activity | Upgrade `@jaggerxtrm/specialists` package. Post-drs41.1, supervisor + `sp stop` invoke `aggregateJobMetricsBestEffort` on every terminal status, so the table fills under normal operation. Backfill historical with `sp db extract --all-missing`. |
|
|
245
|
+
| `sp merge` fails after `sp stop` (older runtime) — "No chain-root job with worktree metadata found" | Upgrade `@jaggerxtrm/specialists` past ofjvj fix. Merge lookup is now DB-first via `readAllJobStatuses()` / `listStatuses()`. Pre-fix workaround was manual `git merge --no-ff feature/<branch>` (skips tsc + conflict gates). |
|
|
246
|
+
| Events about to be pruned but never aggregated | `sp db extract --all-missing` BEFORE `sp db prune --apply`. Prune refuses when extract fails (safe by design). |
|
|
247
|
+
| Emergency: need to prune but extract is wedged | `sp db prune --apply --skip-extract` — raw events deleted without aggregation. Use only when data loss is acceptable. |
|
|
248
|
+
| Historical per-job stats needed | `sp db stats` reads `specialist_job_metrics`. Replaces ad-hoc `status.json` scans. Supports `--format json\|table`. |
|
|
249
|
+
| Pi extensions missing | `specialists init --sync-skills` or reinstall extension registration |
|
|
250
|
+
| Hook config format stale | `specialists init` |
|
|
251
|
+
| Skill symlink / active-skill drift | `specialists init --sync-skills` |
|
|
252
|
+
| Installed default skill differs from canonical package copy | `specialists init --sync-skills` unless local customization is intentional |
|
|
253
|
+
| Skill frontmatter version / synced_at drift | `specialists init --sync-skills` or refresh packaged skills |
|
|
254
|
+
| Mandatory-rules mirror drift (`.specialists/default/mandatory-rules`) | `specialists init --sync-defaults` |
|
|
255
|
+
| `.specialists/mandatory-rules/` overlay present | Leave alone — this is repo overlay, NOT drift |
|
|
256
|
+
| Missing/invalid `template_sets` references | Check all three tiers first; `sp edit <name> --fork-from <base>` then fix references, or sync defaults if mirror missing, or add set to overlay if intended |
|
|
133
257
|
| Unknown manual drift | Stop, inspect, then apply user-approved fix |
|
|
134
258
|
|
|
135
259
|
## Remediation
|
|
136
260
|
|
|
261
|
+
### Fix: Package/runtime version drift
|
|
262
|
+
|
|
263
|
+
If installed npm package version, CLI version, or package root parity checks disagree:
|
|
264
|
+
|
|
265
|
+
```bash
|
|
266
|
+
npm ls @jaggerxtrm/specialists --depth=0
|
|
267
|
+
specialists --version
|
|
268
|
+
sp --version
|
|
269
|
+
```
|
|
270
|
+
|
|
271
|
+
If versions do not align, reinstall or upgrade the package first. After runtime
|
|
272
|
+
version is correct, re-run `specialists init` / sync commands to repair repo drift.
|
|
273
|
+
|
|
137
274
|
### Fix: Specialist configs drifted
|
|
138
275
|
|
|
139
|
-
If `sp doctor
|
|
140
|
-
mismatch:
|
|
276
|
+
If `sp doctor`, JSON validation, or direct diff against package canonical defaults
|
|
277
|
+
shows missing fields, wrong names, or schema mismatch:
|
|
141
278
|
|
|
142
279
|
```bash
|
|
143
|
-
|
|
280
|
+
specialists init --sync-defaults
|
|
144
281
|
```
|
|
145
282
|
|
|
283
|
+
`--sync-defaults` refreshes specialists + mandatory-rules + nodes mirrors.
|
|
284
|
+
|
|
146
285
|
If one specialist needs a small repair and `sp edit` supports it, prefer that over
|
|
147
|
-
full sync.
|
|
286
|
+
full sync. If target specialist lives in default/package layer, fork first:
|
|
287
|
+
|
|
288
|
+
```bash
|
|
289
|
+
sp edit <name> --fork-from <base>
|
|
290
|
+
```
|
|
148
291
|
|
|
149
292
|
### Fix: Hooks not firing
|
|
150
293
|
|
|
151
|
-
If hooks are missing, wrong events,
|
|
294
|
+
If hooks are missing, wrong events, stale script paths, or hook files differ from
|
|
295
|
+
installed package canonical copies:
|
|
152
296
|
|
|
153
297
|
```bash
|
|
154
|
-
|
|
298
|
+
specialists init
|
|
155
299
|
```
|
|
156
300
|
|
|
157
301
|
If runtime exposes a narrower hook sync command, prefer it. Use full init only
|
|
@@ -186,12 +330,55 @@ If doctor reports DB version mismatch or recovery issue:
|
|
|
186
330
|
2. Apply runtime migration command if available.
|
|
187
331
|
3. If no automated migration exists, flag manual intervention.
|
|
188
332
|
|
|
333
|
+
For additive schema bumps (e.g. drs41.1 added `active_runtime_ms` / `waiting_ms` columns within v11): just open any `sp` command in the repo. `initSchema()` runs every migrate-up function on every startup; `migrateToV11` is idempotent — it detects existing v11 schema and `ALTER TABLE ADD COLUMN`s the missing fields. Existing rows are preserved (new fields = NULL until next aggregate). No data loss, no manual SQL.
|
|
334
|
+
|
|
335
|
+
`sp init` and `sp init --sync-defaults` skip the DB entirely when it exists — the only way to wipe `.specialists/db/observability.db` is to delete the file manually.
|
|
336
|
+
|
|
337
|
+
### Fix: metrics aggregation missing or stale
|
|
338
|
+
|
|
339
|
+
Schema v11 introduced `specialist_job_metrics` (aggregated per-job stats). If you see low `metrics_coverage` in the detection output, or want historical stats before running `sp db prune`:
|
|
340
|
+
|
|
341
|
+
```bash
|
|
342
|
+
# Backfill metrics for any job whose events still exist but lack a metrics row.
|
|
343
|
+
sp db extract --all-missing
|
|
344
|
+
|
|
345
|
+
# Inspect specific job metrics.
|
|
346
|
+
sp db extract --job <job-id>
|
|
347
|
+
|
|
348
|
+
# Query aggregates.
|
|
349
|
+
sp db stats
|
|
350
|
+
sp db stats --spec executor --since 7d --format json
|
|
351
|
+
sp db stats --model 'openai-codex/*' --since 30d
|
|
352
|
+
```
|
|
353
|
+
|
|
354
|
+
`sp db prune --apply` automatically extracts for every job whose events will be deleted (unless `--skip-extract`). If extract throws, prune aborts — investigate the failing job instead of bypassing.
|
|
355
|
+
|
|
356
|
+
Safe order before a retention cleanup:
|
|
357
|
+
1. `sp db extract --all-missing` — verify no extract errors.
|
|
358
|
+
2. `sp db prune --before 30d --dry-run` — confirm scope.
|
|
359
|
+
3. `sp db prune --before 30d --apply` — prune with pre-extract built in.
|
|
360
|
+
4. `sp db vacuum` — compact file size.
|
|
361
|
+
|
|
362
|
+
### Fix: Skills/defaults differ from shipped package copy
|
|
363
|
+
|
|
364
|
+
If diff against the installed package shows `.specialists/default/`,
|
|
365
|
+
`.xtrm/skills/default/`, or `.claude/hooks/` drift from shipped canonical files:
|
|
366
|
+
|
|
367
|
+
- If drift is intentional project customization, report it and do not overwrite silently.
|
|
368
|
+
- If drift is unintentional, use the narrowest sync that fixes the affected area:
|
|
369
|
+
|
|
370
|
+
```bash
|
|
371
|
+
specialists init --sync-defaults
|
|
372
|
+
specialists init --sync-skills
|
|
373
|
+
specialists init
|
|
374
|
+
```
|
|
375
|
+
|
|
189
376
|
### Fix: Pi extensions not registered
|
|
190
377
|
|
|
191
378
|
If `quality-gates`, `pi-gitnexus`, or `pi-serena-tools` are missing:
|
|
192
379
|
|
|
193
380
|
```bash
|
|
194
|
-
|
|
381
|
+
specialists init --sync-skills
|
|
195
382
|
```
|
|
196
383
|
|
|
197
384
|
If project uses different extension packaging, re-run install step that writes
|
|
@@ -204,9 +391,9 @@ After fixes, confirm canonical state restored.
|
|
|
204
391
|
```bash
|
|
205
392
|
sp doctor
|
|
206
393
|
sp status
|
|
207
|
-
|
|
208
|
-
|
|
209
|
-
|
|
394
|
+
npm ls @jaggerxtrm/specialists --depth=0 2>/dev/null || true
|
|
395
|
+
specialists --version 2>/dev/null || true
|
|
396
|
+
sp --version 2>/dev/null || true
|
|
210
397
|
|
|
211
398
|
node -e "const fs=require('fs'); const p='.claude/settings.json'; const s=JSON.parse(fs.readFileSync(p,'utf8')); console.log(Boolean(s.hooks || Object.keys(s).length))"
|
|
212
399
|
```
|
|
@@ -214,9 +401,11 @@ node -e "const fs=require('fs'); const p='.claude/settings.json'; const s=JSON.p
|
|
|
214
401
|
Expected outcome:
|
|
215
402
|
- `sp doctor` clean
|
|
216
403
|
- `sp status` no drift / no repair hints
|
|
217
|
-
- `sp` and `specialists` reachable
|
|
404
|
+
- `sp` and `specialists` reachable (`sp` is shorthand; `specialists` is canonical)
|
|
405
|
+
- installed package / CLI versions aligned
|
|
218
406
|
- specialist JSON files valid
|
|
219
|
-
-
|
|
407
|
+
- repo defaults match installed package defaults (or intentional custom drift acknowledged)
|
|
408
|
+
- hooks present on required events and canonical hook files match installed package copy
|
|
220
409
|
- no orphaned worktrees
|
|
221
410
|
- SQLite state healthy
|
|
222
411
|
|
|
@@ -230,6 +419,8 @@ Flag these when automatic fix is unsafe or impossible:
|
|
|
230
419
|
- schema mismatch with no available migration path
|
|
231
420
|
- worktree cleanup would remove user changes
|
|
232
421
|
- extensions required by project are not installed at package level
|
|
422
|
+
- package root is unavailable, so package-vs-installed diff cannot be computed
|
|
423
|
+
- repo intentionally diverges from canonical package copies and user must preserve customizations
|
|
233
424
|
|
|
234
425
|
When manual intervention needed, report:
|
|
235
426
|
1. exact drift
|
|
@@ -245,9 +436,10 @@ After detection + remediation, answer with compact status:
|
|
|
245
436
|
## specialists update complete
|
|
246
437
|
|
|
247
438
|
✓ sp doctor clean
|
|
439
|
+
✓ package / CLI versions aligned
|
|
248
440
|
✓ specialist configs valid
|
|
249
441
|
✓ hooks wired
|
|
250
|
-
✓
|
|
442
|
+
✓ canonical package parity checked
|
|
251
443
|
✓ jobs/worktrees clean
|
|
252
444
|
✓ SQLite healthy
|
|
253
445
|
✓ extensions registered
|
|
@@ -74,6 +74,31 @@ node -e "const s=require('./.pi/settings.json'); console.log(s.skills)" 2>/dev/n
|
|
|
74
74
|
for f in .xtrm/skills/active/*; do [ -L "$f" ] || echo "NOT A SYMLINK: $f"; done
|
|
75
75
|
```
|
|
76
76
|
|
|
77
|
+
## Implementation Self-Check
|
|
78
|
+
|
|
79
|
+
Do not trust the surface commands alone. Before claiming that `xt init` handles
|
|
80
|
+
drift correctly, verify the underlying implementation behavior in the CLI source.
|
|
81
|
+
|
|
82
|
+
Required checks:
|
|
83
|
+
|
|
84
|
+
| File | What to verify |
|
|
85
|
+
|------|----------------|
|
|
86
|
+
| `cli/src/core/drift.ts` | Drift is classified by comparing installed user file hashes against registry hashes from the package payload |
|
|
87
|
+
| `cli/src/core/registry-scaffold.ts` | Drifted files are reported and skipped by default unless `force` is enabled |
|
|
88
|
+
| `cli/src/commands/init.ts` | `xt init` calls the registry install step with `force: false` |
|
|
89
|
+
|
|
90
|
+
What you must confirm from code before reporting success:
|
|
91
|
+
|
|
92
|
+
- `xt init` does check for local drift between the user's `.xtrm` files and the
|
|
93
|
+
package payload that bootstrapped them.
|
|
94
|
+
- That check is hash-based for registry-managed `.xtrm` files, not just a loose
|
|
95
|
+
status heuristic.
|
|
96
|
+
- `xt init -y` is non-destructive for drifted `.xtrm` files by default. It
|
|
97
|
+
preserves local edits unless a separate force path is used.
|
|
98
|
+
|
|
99
|
+
If the implementation no longer matches those rules, stop and report the mismatch
|
|
100
|
+
instead of repeating this skill's older assumptions.
|
|
101
|
+
|
|
77
102
|
## Remediation
|
|
78
103
|
|
|
79
104
|
Two commands cover almost all drift. Know which fixes what:
|
|
@@ -86,6 +111,8 @@ Two commands cover almost all drift. Know which fixes what:
|
|
|
86
111
|
### Fix: Skills symlink stale or active/ view wrong
|
|
87
112
|
|
|
88
113
|
`xt claude install` does NOT rebuild skills. Only `xt init` does (Phase 6b).
|
|
114
|
+
`xt init -y` will repair missing/outdated registry-managed files, but it will
|
|
115
|
+
preserve locally drifted `.xtrm` files by default.
|
|
89
116
|
|
|
90
117
|
```bash
|
|
91
118
|
xt init -y
|
|
@@ -159,6 +186,13 @@ node -e "const s=require('./.pi/settings.json'); console.log(s.skills.includes('
|
|
|
159
186
|
# Must output: true
|
|
160
187
|
```
|
|
161
188
|
|
|
189
|
+
Also restate the implementation-level conclusion in your report:
|
|
190
|
+
|
|
191
|
+
- `xt init` verified drift against package registry hashes
|
|
192
|
+
- local drifted `.xtrm` files were preserved by default
|
|
193
|
+
- no forced overwrite path was used unless explicitly requested
|
|
194
|
+
|
|
195
|
+
|
|
162
196
|
If `xt status` still shows drift after targeted fixes, run the full sync:
|
|
163
197
|
```bash
|
|
164
198
|
xt init
|
|
@@ -0,0 +1,150 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: using-kpi
|
|
3
|
+
description: >-
|
|
4
|
+
Analyze specialist KPI data in observability SQLite. Use for runtime, payload,
|
|
5
|
+
waiting, tool-call, and outlier analysis. Token estimates use cl100k_base-style
|
|
6
|
+
approximation with ~±5% accuracy.
|
|
7
|
+
gemini-command: using-kpi
|
|
8
|
+
version: 3.1.0
|
|
9
|
+
---
|
|
10
|
+
|
|
11
|
+
# using-kpi
|
|
12
|
+
|
|
13
|
+
KPI analysis skill for `sp db stats` / `sp db extract` data.
|
|
14
|
+
|
|
15
|
+
## Quick rule
|
|
16
|
+
|
|
17
|
+
`active_runtime_ms` = real paid runtime. Rank by that first. `elapsed_ms` is total wall time. `waiting_ms` catches forgotten keep-alives.
|
|
18
|
+
|
|
19
|
+
Token counts are approximate, cl100k_base-style, about ±5%. Bytes are exact UTF-8 size.
|
|
20
|
+
|
|
21
|
+
## Recipe 1 — specialist × model leaderboard by active cost
|
|
22
|
+
|
|
23
|
+
```bash
|
|
24
|
+
sp db stats --format json \
|
|
25
|
+
| jq -r '
|
|
26
|
+
.rows
|
|
27
|
+
| group_by([.specialist, .model])
|
|
28
|
+
| map({
|
|
29
|
+
specialist: .[0].specialist,
|
|
30
|
+
model: .[0].model,
|
|
31
|
+
jobs: length,
|
|
32
|
+
active_ms: (map((.active_runtime_ms // 0)) | add),
|
|
33
|
+
total_ms: (map((.total_runtime_ms // .elapsed_ms // 0)) | add),
|
|
34
|
+
turns: (map((.total_turns // 0)) | add),
|
|
35
|
+
tools: (map((.total_tools // 0)) | add),
|
|
36
|
+
payload_kb: (map((.payload_kb // 0)) | add)
|
|
37
|
+
})
|
|
38
|
+
| sort_by(-.active_ms, -.jobs)
|
|
39
|
+
| .[]
|
|
40
|
+
| [ .specialist, .model, .jobs, .active_ms, .total_ms, .turns, .tools, .payload_kb ]
|
|
41
|
+
| @tsv'
|
|
42
|
+
```
|
|
43
|
+
|
|
44
|
+
## Recipe 2 — outliers above p95
|
|
45
|
+
|
|
46
|
+
```bash
|
|
47
|
+
sp db stats --format json \
|
|
48
|
+
| jq '
|
|
49
|
+
.rows as $rows
|
|
50
|
+
| {
|
|
51
|
+
active: ($rows | map(.active_runtime_ms // 0) | sort),
|
|
52
|
+
tools: ($rows | map(.total_tools // 0) | sort),
|
|
53
|
+
turns: ($rows | map(.total_turns // 0) | sort),
|
|
54
|
+
payload: ($rows | map(.payload_kb // 0) | sort)
|
|
55
|
+
} as $s
|
|
56
|
+
| {
|
|
57
|
+
active_p95: $s.active[(($s.active|length)*95/100|floor)],
|
|
58
|
+
tools_p95: $s.tools[(($s.tools|length)*95/100|floor)],
|
|
59
|
+
turns_p95: $s.turns[(($s.turns|length)*95/100|floor)],
|
|
60
|
+
payload_p95: $s.payload[(($s.payload|length)*95/100|floor)]
|
|
61
|
+
} as $p
|
|
62
|
+
| $rows
|
|
63
|
+
| map(select(
|
|
64
|
+
((.active_runtime_ms // 0) >= $p.active_p95) or
|
|
65
|
+
((.total_tools // 0) >= $p.tools_p95) or
|
|
66
|
+
((.total_turns // 0) >= $p.turns_p95) or
|
|
67
|
+
((.payload_kb // 0) >= $p.payload_p95)
|
|
68
|
+
))
|
|
69
|
+
| .[]
|
|
70
|
+
| [ .job_id, .specialist, .model, .active_runtime_ms, .total_tools, .total_turns, .payload_kb ]
|
|
71
|
+
| @tsv'
|
|
72
|
+
```
|
|
73
|
+
|
|
74
|
+
## Recipe 3 — payload bloat ranking
|
|
75
|
+
|
|
76
|
+
```bash
|
|
77
|
+
sp db stats --with-payload --format json \
|
|
78
|
+
| jq -r '
|
|
79
|
+
.rows
|
|
80
|
+
| group_by(.specialist)
|
|
81
|
+
| map({
|
|
82
|
+
specialist: .[0].specialist,
|
|
83
|
+
jobs: length,
|
|
84
|
+
avg_payload_kb: ((map((.payload_kb // 0)) | add) / length),
|
|
85
|
+
max_payload_kb: (map((.payload_kb // 0)) | max)
|
|
86
|
+
})
|
|
87
|
+
| sort_by(-.avg_payload_kb)
|
|
88
|
+
| .[:10]
|
|
89
|
+
| .[]
|
|
90
|
+
| [ .specialist, .jobs, (.avg_payload_kb|tostring), (.max_payload_kb|tostring) ]
|
|
91
|
+
| @tsv'
|
|
92
|
+
```
|
|
93
|
+
|
|
94
|
+
## Recipe 4 — waiting-state hygiene
|
|
95
|
+
|
|
96
|
+
```bash
|
|
97
|
+
sp db stats --format json \
|
|
98
|
+
| jq -r '
|
|
99
|
+
.rows
|
|
100
|
+
| map(select((.waiting_s? // 0) != 0))
|
|
101
|
+
| map(. + {waiting_ratio: ((.waiting_ms // 0) / ((.total_runtime_ms // .elapsed_ms // 1) + 0.0))})
|
|
102
|
+
| sort_by(-.waiting_ratio, -.waiting_ms)
|
|
103
|
+
| .[]
|
|
104
|
+
| [ .job_id, .specialist, .model, (.waiting_ms|tostring), (.total_runtime_ms // .elapsed_ms|tostring), (.waiting_ratio|tostring) ]
|
|
105
|
+
| @tsv'
|
|
106
|
+
```
|
|
107
|
+
|
|
108
|
+
## Recipe 5 — tool-call distribution per specialist
|
|
109
|
+
|
|
110
|
+
```bash
|
|
111
|
+
sp db stats --format json \
|
|
112
|
+
| jq -r '
|
|
113
|
+
.rows
|
|
114
|
+
| group_by(.specialist)
|
|
115
|
+
| map({
|
|
116
|
+
specialist: .[0].specialist,
|
|
117
|
+
counts: (map(.tool_call_counts_json? // "{}")
|
|
118
|
+
| map(fromjson)
|
|
119
|
+
| add)
|
|
120
|
+
})
|
|
121
|
+
| .[]
|
|
122
|
+
| .counts
|
|
123
|
+
| to_entries
|
|
124
|
+
| sort_by(-.value)
|
|
125
|
+
| .[]
|
|
126
|
+
| [ .key, .value ]
|
|
127
|
+
| @tsv'
|
|
128
|
+
```
|
|
129
|
+
|
|
130
|
+
## Recipe 6 — payload vs active runtime correlation
|
|
131
|
+
|
|
132
|
+
```bash
|
|
133
|
+
sp db stats --with-payload --format json \
|
|
134
|
+
| jq -r '
|
|
135
|
+
.rows
|
|
136
|
+
| map(select((.payload_kb? // 0) > 0 and ((.active_runtime_ms? // 0) > 0)))
|
|
137
|
+
| map([(.payload_kb|tonumber), (.active_runtime_ms|tonumber)])
|
|
138
|
+
| if length < 2 then empty else
|
|
139
|
+
(map(.[0]) | add / length) as $mx |
|
|
140
|
+
(map(.[1]) | add / length) as $my |
|
|
141
|
+
(map((.[0]-$mx)*(.[1]-$my)) | add) /
|
|
142
|
+
((map((.[0]-$mx)^2) | add) * (map((.[1]-$my)^2) | add)) ^ 0.5
|
|
143
|
+
end'
|
|
144
|
+
```
|
|
145
|
+
|
|
146
|
+
## References
|
|
147
|
+
|
|
148
|
+
- `docs/observability-metrics.md`
|
|
149
|
+
- `src/cli/db.ts`
|
|
150
|
+
- `src/specialist/observability-sqlite.ts`
|