@codyswann/lisa 2.237.0 → 2.239.0
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/dist/core/learnings-writer.d.ts +22 -0
- package/dist/core/learnings-writer.d.ts.map +1 -1
- package/dist/core/learnings-writer.js +51 -8
- package/dist/core/learnings-writer.js.map +1 -1
- package/dist/core/learnings.d.ts +1 -1
- package/dist/core/learnings.d.ts.map +1 -1
- package/dist/core/learnings.js +1 -1
- package/dist/core/learnings.js.map +1 -1
- package/package.json +1 -1
- package/plugins/lisa/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa/.codex-plugin/plugin.json +1 -1
- package/plugins/lisa/.codex-plugin/skills/lisa-atlassian-access/SKILL.md +12 -1
- package/plugins/lisa/.codex-plugin/skills/lisa-drive-pr-to-merge/SKILL.md +62 -8
- package/plugins/lisa/.codex-plugin/skills/lisa-git-submit-pr/SKILL.md +3 -2
- package/plugins/lisa/.codex-plugin/skills/lisa-github-build-intake/SKILL.md +5 -1
- package/plugins/lisa/.codex-plugin/skills/lisa-implement/SKILL.md +2 -0
- package/plugins/lisa/.codex-plugin/skills/lisa-jira-build-intake/SKILL.md +5 -1
- package/plugins/lisa/.codex-plugin/skills/lisa-linear-access/SKILL.md +54 -0
- package/plugins/lisa/.codex-plugin/skills/lisa-linear-build-intake/SKILL.md +5 -1
- package/plugins/lisa/.codex-plugin/skills/lisa-persist-learning/SKILL.md +145 -0
- package/plugins/lisa/.codex-plugin/skills/lisa-persist-learning/agents/openai.yaml +4 -0
- package/plugins/lisa/agents/learning-judge.md +119 -0
- package/plugins/lisa/commands/persist-learning.md +6 -0
- package/plugins/lisa/rules/eager/rejection-detection.md +40 -0
- package/plugins/lisa/rules/reference/rejection-detection.md +103 -0
- package/plugins/lisa/skills/lisa-atlassian-access/SKILL.md +12 -1
- package/plugins/lisa/skills/lisa-drive-pr-to-merge/SKILL.md +62 -8
- package/plugins/lisa/skills/lisa-git-submit-pr/SKILL.md +3 -2
- package/plugins/lisa/skills/lisa-github-build-intake/SKILL.md +5 -1
- package/plugins/lisa/skills/lisa-implement/SKILL.md +2 -0
- package/plugins/lisa/skills/lisa-jira-build-intake/SKILL.md +5 -1
- package/plugins/lisa/skills/lisa-linear-access/SKILL.md +54 -0
- package/plugins/lisa/skills/lisa-linear-build-intake/SKILL.md +5 -1
- package/plugins/lisa/skills/lisa-persist-learning/SKILL.md +145 -0
- package/plugins/lisa/skills/lisa-persist-learning/agents/openai.yaml +4 -0
- package/plugins/lisa-agy/agents/learning-judge.md +119 -0
- package/plugins/lisa-agy/commands/lisa/persist-learning.md +6 -0
- package/plugins/lisa-agy/plugin.json +1 -1
- package/plugins/lisa-agy/skills/lisa-atlassian-access/SKILL.md +12 -1
- package/plugins/lisa-agy/skills/lisa-drive-pr-to-merge/SKILL.md +62 -8
- package/plugins/lisa-agy/skills/lisa-git-submit-pr/SKILL.md +3 -2
- package/plugins/lisa-agy/skills/lisa-github-build-intake/SKILL.md +5 -1
- package/plugins/lisa-agy/skills/lisa-implement/SKILL.md +2 -0
- package/plugins/lisa-agy/skills/lisa-jira-build-intake/SKILL.md +5 -1
- package/plugins/lisa-agy/skills/lisa-linear-access/SKILL.md +54 -0
- package/plugins/lisa-agy/skills/lisa-linear-build-intake/SKILL.md +5 -1
- package/plugins/lisa-agy/skills/lisa-persist-learning/SKILL.md +145 -0
- package/plugins/lisa-cdk/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-cdk/.codex-plugin/plugin.json +1 -1
- package/plugins/lisa-cdk-agy/plugin.json +1 -1
- package/plugins/lisa-cdk-copilot/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-cdk-cursor/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-copilot/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-copilot/agents/learning-judge.agent.md +119 -0
- package/plugins/lisa-copilot/commands/lisa/persist-learning.md +6 -0
- package/plugins/lisa-copilot/rules/eager/rejection-detection.md +40 -0
- package/plugins/lisa-copilot/rules/reference/rejection-detection.md +103 -0
- package/plugins/lisa-copilot/skills/lisa-atlassian-access/SKILL.md +12 -1
- package/plugins/lisa-copilot/skills/lisa-drive-pr-to-merge/SKILL.md +62 -8
- package/plugins/lisa-copilot/skills/lisa-git-submit-pr/SKILL.md +3 -2
- package/plugins/lisa-copilot/skills/lisa-github-build-intake/SKILL.md +5 -1
- package/plugins/lisa-copilot/skills/lisa-implement/SKILL.md +2 -0
- package/plugins/lisa-copilot/skills/lisa-jira-build-intake/SKILL.md +5 -1
- package/plugins/lisa-copilot/skills/lisa-linear-access/SKILL.md +54 -0
- package/plugins/lisa-copilot/skills/lisa-linear-build-intake/SKILL.md +5 -1
- package/plugins/lisa-copilot/skills/lisa-persist-learning/SKILL.md +145 -0
- package/plugins/lisa-cursor/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-cursor/agents/learning-judge.md +119 -0
- package/plugins/lisa-cursor/commands/lisa/persist-learning.md +6 -0
- package/plugins/lisa-cursor/rules/rejection-detection-reference.mdc +108 -0
- package/plugins/lisa-cursor/rules/rejection-detection.mdc +45 -0
- package/plugins/lisa-cursor/skills/lisa-atlassian-access/SKILL.md +12 -1
- package/plugins/lisa-cursor/skills/lisa-drive-pr-to-merge/SKILL.md +62 -8
- package/plugins/lisa-cursor/skills/lisa-git-submit-pr/SKILL.md +3 -2
- package/plugins/lisa-cursor/skills/lisa-github-build-intake/SKILL.md +5 -1
- package/plugins/lisa-cursor/skills/lisa-implement/SKILL.md +2 -0
- package/plugins/lisa-cursor/skills/lisa-jira-build-intake/SKILL.md +5 -1
- package/plugins/lisa-cursor/skills/lisa-linear-access/SKILL.md +54 -0
- package/plugins/lisa-cursor/skills/lisa-linear-build-intake/SKILL.md +5 -1
- package/plugins/lisa-cursor/skills/lisa-persist-learning/SKILL.md +145 -0
- package/plugins/lisa-expo/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-expo/.codex-plugin/plugin.json +1 -1
- package/plugins/lisa-expo-agy/plugin.json +1 -1
- package/plugins/lisa-expo-copilot/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-expo-cursor/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-harper-fabric/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-harper-fabric/.codex-plugin/plugin.json +1 -1
- package/plugins/lisa-harper-fabric-agy/plugin.json +1 -1
- package/plugins/lisa-harper-fabric-copilot/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-harper-fabric-cursor/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-nestjs/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-nestjs/.codex-plugin/plugin.json +1 -1
- package/plugins/lisa-nestjs-agy/plugin.json +1 -1
- package/plugins/lisa-nestjs-copilot/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-nestjs-cursor/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-openclaw/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-openclaw/.codex-plugin/plugin.json +1 -1
- package/plugins/lisa-openclaw-agy/plugin.json +1 -1
- package/plugins/lisa-openclaw-copilot/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-openclaw-cursor/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-phaser/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-phaser/.codex-plugin/plugin.json +1 -1
- package/plugins/lisa-phaser-agy/plugin.json +1 -1
- package/plugins/lisa-phaser-copilot/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-phaser-cursor/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-rails/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-rails/.codex-plugin/plugin.json +1 -1
- package/plugins/lisa-rails-agy/plugin.json +1 -1
- package/plugins/lisa-rails-copilot/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-rails-cursor/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-typescript/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-typescript/.codex-plugin/plugin.json +1 -1
- package/plugins/lisa-typescript-agy/plugin.json +1 -1
- package/plugins/lisa-typescript-copilot/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-typescript-cursor/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-wiki/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-wiki/.codex-plugin/plugin.json +1 -1
- package/plugins/lisa-wiki-agy/plugin.json +1 -1
- package/plugins/lisa-wiki-copilot/.claude-plugin/plugin.json +1 -1
- package/plugins/lisa-wiki-cursor/.claude-plugin/plugin.json +1 -1
- package/plugins/src/base/agents/learning-judge.md +119 -0
- package/plugins/src/base/commands/persist-learning.md +6 -0
- package/plugins/src/base/rules/eager/rejection-detection.md +40 -0
- package/plugins/src/base/rules/reference/rejection-detection.md +103 -0
- package/plugins/src/base/skills/lisa-atlassian-access/SKILL.md +12 -1
- package/plugins/src/base/skills/lisa-drive-pr-to-merge/SKILL.md +62 -8
- package/plugins/src/base/skills/lisa-git-submit-pr/SKILL.md +3 -2
- package/plugins/src/base/skills/lisa-github-build-intake/SKILL.md +5 -1
- package/plugins/src/base/skills/lisa-implement/SKILL.md +2 -0
- package/plugins/src/base/skills/lisa-jira-build-intake/SKILL.md +5 -1
- package/plugins/src/base/skills/lisa-linear-access/SKILL.md +54 -0
- package/plugins/src/base/skills/lisa-linear-build-intake/SKILL.md +5 -1
- package/plugins/src/base/skills/lisa-persist-learning/SKILL.md +145 -0
|
@@ -0,0 +1,145 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: lisa-persist-learning
|
|
3
|
+
description: This skill should be used when a candidate learning (from a failure signal, rejection, or debrief) needs to be judged and routed. It computes a stable fingerprint, runs the candidate through the hostile-default learning-judge gate, and performs exactly the verdict's side effects — a dropped-with-reason note on the triggering issue (drop), an upstream handoff marker (lisa-upstream), or a confidence-routed pull request that touches only the learnings surface (durable-learning). Idempotent via marker dedupe; headless-safe; never blocks the primary build flow.
|
|
4
|
+
---
|
|
5
|
+
|
|
6
|
+
# Persist Learning
|
|
7
|
+
|
|
8
|
+
Route ONE candidate learning through the judgment gate and act on the verdict. Candidate: $ARGUMENTS
|
|
9
|
+
|
|
10
|
+
Most candidates are dropped — that is the gate working, not a failure. Nothing is ever silent (every drop leaves a visible note), and no learning content ever reaches the learnings surface outside a pull request.
|
|
11
|
+
|
|
12
|
+
## Candidate Input
|
|
13
|
+
|
|
14
|
+
Accept the candidate as JSON or `key=value` fields:
|
|
15
|
+
|
|
16
|
+
- `rule` — the proposed learning (must fit the executable contract: ≤240 chars, ≤2 lines)
|
|
17
|
+
- `why` — the causal claim
|
|
18
|
+
- `provenance` — stable refs (issues/PRs/commits/comments), ≤20
|
|
19
|
+
- `evidence_links` — concrete evidence refs available for citation
|
|
20
|
+
- `scope_hint` — `project` | `upstream` (a hint; the judge decides)
|
|
21
|
+
- `triggering_issue` — the issue/work item whose failure produced this candidate (required)
|
|
22
|
+
- `fingerprint` — optional; computed below when absent
|
|
23
|
+
|
|
24
|
+
## Phase 0 — Fingerprint (stable dedupe key)
|
|
25
|
+
|
|
26
|
+
Every marker and branch below keys off one deterministic fingerprint:
|
|
27
|
+
|
|
28
|
+
```text
|
|
29
|
+
fingerprint = "sll4-" + first 12 hex chars of sha1(normalized_rule + "\n" + triggering_issue)
|
|
30
|
+
normalized_rule = rule lowercased, all whitespace runs collapsed to single spaces, trimmed
|
|
31
|
+
```
|
|
32
|
+
|
|
33
|
+
```bash
|
|
34
|
+
NORM=$(printf '%s' "$RULE" | tr '[:upper:]' '[:lower:]' | tr -s '[:space:]' ' ' | sed 's/^ *//; s/ *$//')
|
|
35
|
+
FP="sll4-$(printf '%s\n%s' "$NORM" "$TRIGGERING_ISSUE" | shasum -a 1 | cut -c1-12)"
|
|
36
|
+
```
|
|
37
|
+
|
|
38
|
+
(Use `sha1sum` where `shasum` is unavailable.) The same rule for the same triggering issue always produces the same fingerprint, so re-runs dedupe instead of duplicating comments, PRs, or branches.
|
|
39
|
+
|
|
40
|
+
## Phase 1 — Judge (mandatory, never skipped)
|
|
41
|
+
|
|
42
|
+
Invoke the `learning-judge` agent (via the Agent/Task tool with `subagent_type: "learning-judge"` — the same invoke pattern `learner` uses for `skill-evaluator`) with the full candidate including the fingerprint. It returns a verdict: `classification`, `cited_evidence[]`, `rationale`, `confidence` (durable only), `disposition`.
|
|
43
|
+
|
|
44
|
+
**Respect the verdict — do not override it.** Never re-run the judge hoping for a different answer, and never persist anything the judge did not classify `durable-learning`.
|
|
45
|
+
|
|
46
|
+
## Phase 2 — Route by disposition
|
|
47
|
+
|
|
48
|
+
All issue/PR comments below follow the marker-dedupe discipline from `lisa-github-write-prd` Phase 2, each with its own producer tag: match on the **marker, never the title or text**; **exactly one marker per body**; **never write a markerless body** (it breaks all future dedupe); include the eventual-consistency guard — when the `gh` search index is stale, also enumerate the bodies directly (`gh issue view <n> --json comments --jq '.comments[].body'` or `gh pr list --json number,body`) and grep for the marker before deciding to create.
|
|
49
|
+
|
|
50
|
+
### `drop` (classification `one-off` or `misunderstanding/spec-gap`)
|
|
51
|
+
|
|
52
|
+
Post **one** comment on the triggering issue and write **nothing** — zero bytes — to the learnings surface (not a stub, not a placeholder):
|
|
53
|
+
|
|
54
|
+
```markdown
|
|
55
|
+
<!-- [lisa-learning-drop] key=<fingerprint> -->
|
|
56
|
+
Dropped (<classification> — <plain-language gloss>): <reason>.
|
|
57
|
+
```
|
|
58
|
+
|
|
59
|
+
The note is one line naming the classification (with its fixed plain-language gloss) and the reason, readable by a non-technical operator. Use exactly these glosses per class:
|
|
60
|
+
|
|
61
|
+
| Classification | Gloss |
|
|
62
|
+
|----------------|-------|
|
|
63
|
+
| `one-off` | a one-time fluke, not a recurring pattern |
|
|
64
|
+
| `misunderstanding/spec-gap` | traced to an unclear requirement, not a durable lesson |
|
|
65
|
+
| `lisa-upstream` | root cause is Lisa itself; routed upstream |
|
|
66
|
+
|
|
67
|
+
(The `lisa-upstream` gloss is used by the handoff note below, not by a drop note.) Dedupe before posting: if any comment on the triggering issue already carries `[lisa-learning-drop] key=<fingerprint>`, do not post again — report the existing note.
|
|
68
|
+
|
|
69
|
+
### `handoff-upstream` (classification `lisa-upstream`)
|
|
70
|
+
|
|
71
|
+
Emit the handoff marker only — file **nothing** (no upstream issue, no local rule; the upstream filing flow [SLL-5] consumes this marker later):
|
|
72
|
+
|
|
73
|
+
```markdown
|
|
74
|
+
<!-- [lisa-learning-upstream-handoff] key=<fingerprint> -->
|
|
75
|
+
Upstream handoff (lisa-upstream — root cause is Lisa itself; routed upstream): <reason>. Nothing persisted locally; upstream filing is a separate flow.
|
|
76
|
+
```
|
|
77
|
+
|
|
78
|
+
Same one-comment marker dedupe on the triggering issue.
|
|
79
|
+
|
|
80
|
+
### `persist` (classification `durable-learning`)
|
|
81
|
+
|
|
82
|
+
Continue to Phase 3.
|
|
83
|
+
|
|
84
|
+
## Phase 3 — Persist via PR (durable-learning only)
|
|
85
|
+
|
|
86
|
+
No learning content is ever committed without a PR — there is no other write path, and the PR must touch **only** the learnings surface (any other changed file is a bug).
|
|
87
|
+
|
|
88
|
+
1. **PR dedupe.** Search all PRs for the marker `[lisa-learning-pr] key=<fingerprint>` in the body (`gh pr list --state all --search '"<marker>" in:body' --json number,url`), with the stale-index guard above. If one exists, reference it and stop — never open a duplicate.
|
|
89
|
+
2. **Resolve the learnings surface path — never hardcode it.** The canonical path is `resolveProjectLearningsFile` from `@codyswann/lisa/learnings`: the `PROJECT_LEARNINGS.md` sibling of the configured `.lisa.config.json` `projectRulesFile` (default `.claude/rules/PROJECT_LEARNINGS.md`):
|
|
90
|
+
|
|
91
|
+
```bash
|
|
92
|
+
LEARNINGS_FILE=$(node -e 'import("@codyswann/lisa/learnings").then(async m => { const c = await m.readProjectConfig(process.cwd()); console.log(m.resolveProjectLearningsFile(c)); })')
|
|
93
|
+
```
|
|
94
|
+
|
|
95
|
+
3. **Consolidation check (mandatory before writing).** Parse the existing entries (`parseLearningsFile` from `@codyswann/lisa/learnings`) and look for entries related to the new rule (same failure class, overlapping topic, or near-duplicate wording). Then write through the executable contract — **never hand-edit the markdown**:
|
|
96
|
+
- **Related entry found** → consolidate via `persistConsolidatedLearning(projectRoot, entry, { supersede: [<related ids>] })`, merging the old entry's still-true content into the new rule. Never append a near-duplicate sibling — a sibling is a bug that fails review.
|
|
97
|
+
- **No related entry** → append via `persistLearningEntry(projectRoot, entry)` and state in the PR body why appending was correct.
|
|
98
|
+
- Entry mapping: `id` = the fingerprint; `rule`/`why`/`provenance` from the candidate; `first_learned` = `last_confirmed` = today (ISO date; on consolidation keep the superseded entry's earliest `first_learned`); `confidence` = the judge's `high`/`low`. The writer re-asserts the entry and token budgets — an over-budget failure means consolidate harder or drop, never truncate by hand.
|
|
99
|
+
4. **Branch + commit.** Work on branch `learning/<fingerprint>`. Commit only the learnings file; verify with `git diff --name-only` that the diff touches nothing else.
|
|
100
|
+
5. **PR body.** Exactly one marker line plus the reviewable story:
|
|
101
|
+
|
|
102
|
+
```markdown
|
|
103
|
+
<!-- [lisa-learning-pr] key=<fingerprint> -->
|
|
104
|
+
|
|
105
|
+
## Learning
|
|
106
|
+
<rule> — <judge rationale>
|
|
107
|
+
|
|
108
|
+
## Provenance
|
|
109
|
+
- Triggering issue: <ref>
|
|
110
|
+
- Ancestor issue/PR (the past failure this is round 2 of): <ref or none found>
|
|
111
|
+
- Rejection comments that produced the candidate: <refs or none>
|
|
112
|
+
- Cited evidence (from the judge): <refs>
|
|
113
|
+
|
|
114
|
+
## Consolidation decision
|
|
115
|
+
<"Superseded entry id(s) X, Y: <why they merged>" | "Appended: no related entry exists — <one-line search summary>">
|
|
116
|
+
```
|
|
117
|
+
|
|
118
|
+
For a **low-confidence** PR, the body must additionally end with this fixed decision-framing line so the human at the gate knows exactly what merging means:
|
|
119
|
+
|
|
120
|
+
```markdown
|
|
121
|
+
**Merge** to adopt this rule into every future session for all six coding agents; **close** to discard it.
|
|
122
|
+
```
|
|
123
|
+
|
|
124
|
+
6. **Confidence routing.**
|
|
125
|
+
- **`high`** → submit through `lisa-git-submit-pr` with its defaults: auto-merge ON, merging through the project's normal gates.
|
|
126
|
+
- **`low`** → submit through `lisa-git-submit-pr` with `auto_merge=false` (drives the PR to green-and-open `awaiting-human`, never merging — even on repos that disallow auto-merge). Then apply the human-triage label, creating it idempotently first (the `lisa-drive-pr-to-merge` label pattern — `|| true` tolerates only already-exists, so verify):
|
|
127
|
+
|
|
128
|
+
```bash
|
|
129
|
+
gh label create "learning:needs-triage" \
|
|
130
|
+
--description "Low-confidence learning PR awaiting human triage" \
|
|
131
|
+
--color FBCA04 || true
|
|
132
|
+
gh label list --search "learning:needs-triage" --json name \
|
|
133
|
+
--jq '[.[].name] | contains(["learning:needs-triage"])' # must print true
|
|
134
|
+
gh pr edit <pr> --add-label "learning:needs-triage"
|
|
135
|
+
```
|
|
136
|
+
|
|
137
|
+
A low-confidence PR sitting OPEN with green checks and `autoMergeRequest: null` is this mode's **success** state — a human merges or closes it. All pending low-confidence learning PRs are findable by filtering on the `learning:needs-triage` label (`gh pr list --label "learning:needs-triage"`).
|
|
138
|
+
|
|
139
|
+
## Rules
|
|
140
|
+
|
|
141
|
+
- **Headless-safe**: no interactive prompts; must run identically under an intake cron.
|
|
142
|
+
- **Never block the build**: if judging or persistence fails, report the failure and let the primary flow continue — shipping the triggering issue always outranks recording a learning about it.
|
|
143
|
+
- **No learning loops about learning**: never feed this skill a candidate whose triggering artifact is itself learning machinery (a `[lisa-learning-*]`-marked comment, learning PR, or handoff); the judge's Step 0 guard backstops this.
|
|
144
|
+
- **Idempotent**: re-running with the same candidate posts no duplicate comment, opens no duplicate PR, and writes no duplicate entry — the fingerprint and markers guarantee it.
|
|
145
|
+
- **One write path**: the learnings surface changes only through `persistLearningEntry` / `persistConsolidatedLearning` inside a PR. Never hand-edit the file, never commit it to the default branch directly.
|