Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →notque/retro
Learning system interface: stats, search, graduate, clear learnings. Backed by learning.db (SQLite + FTS5).
Passed 15 of 19 test cases with the skill, and 3 of 19 without it.
and 15 more in the full report
14 fixed4 passed either way5 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
We need to search our negative experiment registry in `docs/what-didnt-work.md` for the term 'speculative-decoding'. What exact shell command should be executed?
| Checked | Without skill | With skill |
|---|---|---|
| Executes grep -n -i "speculative-decoding" docs/what-didnt-work.md to search the doc. | Missed | Passed |
The instructions your model reads when the skill loads.
Read-only retrospective front door. Two things to read: docs/what-didnt-work.md, the negative-results registry that records which experiments lost; and the routing and review telemetry in learning.db, queried through scripts/learning-db.py. Both are stores this skill reports on — it records nothing itself. Hooks write the telemetry; humans write the registry.
Parse the user's argument to pick the subcommand. Default to what-didnt-work when no argument is given.
| Argument | Subcommand |
|---|---|
| (none), what-didnt-work, negative results | what-didnt-work |
| routing, route health, route stats | routing |
| reviews, review roi, false positives | reviews |
Print the negative-results registry, the list of experiments that lost. Read it before re-running an experiment so a known-dead path is not retried.
The registry is a doc, not a DB table: docs/what-didnt-work.md is the capture, store, and query target.
Step 1: Read and print the registry.
Use the Read tool on docs/what-didnt-work.md and present it. Group by the dated ## YYYY-MM-DD headings; show each entry's Decision verdict (rejected / deferred / revisit-if) up front so a scan answers "did we already reject this?".
NEGATIVE RESULTS (docs/what-didnt-work.md)
==========================================
## [date] [experiment]
Decision: [rejected | deferred | revisit-if <condition>]
What happened: [one line]
...
If the file is missing, report that no negative results are recorded yet and point the user at the format in CONTRIBUTING.md.
test cases passed, vs 3 / 19 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.