lukaszraczylo/claude-adam
1
0
Fork 0
mirror of https://github.com/lukaszraczylo/claude-adam.git synced 2026-06-09 23:19:12 +00:00
Code Issues Packages Projects Releases Wiki Activity

rubric: lower single-session struggle threshold to >=1 entry

Browse Source 
The hook emits struggle signals only after crossing internal thresholds
(3 retries, 8 tools no-prompt, 4 edits to one file, 2 build failures, etc.).
Each journal entry is therefore meaningful evidence on its own. Old rule
required >=3 entries within single session, which the once-per-thing
emission design rarely produces. New rule: >=1 struggle entry qualifies
for proposal at +2 weight (cross-session bonus does not stack).

Auto-apply still requires cross_session_evidence; single-session-only
proposals always queue for review.
This commit is contained in:
Lukasz Raczylo
2026-05-10 03:08:02 +01:00
parent 8ccd4aa4a3
commit 2b91db6bf3
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
agents/adam.md
+2 -2
View File
@@ -65,7 +65,7 @@ The hook emits these `type` values into the journal:
- `build_loop`: cluster by `session`.
- `subagent_dispatch_pattern`: cluster by `subagent_type`.
6. **Multi-axis correlation**: for each session that produced ≥2 distinct struggle types (`tool_error_loop`, `dead_end`, `weak_agent`, `retry_loop`, `edit_churn`, `build_loop`), tag clusters from that session as `multi_axis: true`. This grants +1 confidence at scoring.
7. For each cluster qualifying under the rubric — ≥3× across ≥2 sessions, OR ≥3× within a single session for struggle types, OR (for `correction`) ≥3 occurrences across ≥2 cwds:
7. For each cluster qualifying under the rubric — ≥3 occurrences across ≥2 sessions, OR (for struggle types `tool_error_loop`, `dead_end`, `weak_agent`, `retry_loop`, `edit_churn`, `build_loop`) ≥1 entry within a single session, OR (for `correction`) ≥3 occurrences across ≥2 cwds:
a. If cluster topic matches a rejected idea (≥2 token overlap with rejection's `# Why`), skip with reason `"rejected-similar"`.
b. Pull ~20 messages of transcript context from `transcripts_root` to enrich. Never read full transcripts.
c. **Solution synthesis** (when type would be `skill_new` AND cluster qualifies for proposal): pull additional ~30 messages of transcript window around the friction events (~50 messages total). Extract:
@@ -147,7 +147,7 @@ When the main thread applies a `skill_new` proposal:
Sum:
- Signal repeated ≥3× across ≥2 sessions: **+2**
- Struggle signal (`tool_error_loop`, `dead_end`, `weak_agent`, `retry_loop`, `edit_churn`, `build_loop`) repeated ≥3× within a single session: **+2** *(does not stack with the cross-session bonus — pick whichever applies, never both)*
- Struggle signal (`tool_error_loop`, `dead_end`, `weak_agent`, `retry_loop`, `edit_churn`, `build_loop`) appearing ≥1× within a single session: **+2** *(each struggle entry already represents a hook-side threshold crossing — e.g. 8 tools without a prompt, 3 same-args retries, 4 edits to one file. Treat each entry as one piece of evidence. Does not stack with the cross-session bonus.)*
- Transcript contains positive endorsement (`yes`, `exactly`, `do that`, `keep doing`) within 2 messages of related action: **+2**
- Multi-axis cluster (≥2 distinct struggle types in same session): **+1**
- Type-bias penalty from feedback loop (≥3 rejections, applied:rejected ratio <1:2 for this `type`): **-1**