ds-ledger
Skill80 starsChanged 15 days agoMITView on GitHub
Keep a one-row-per-decision log during long or unattended runs, and re-read it before each step so the run does not drift from its goal. Use for multi-phase or autonomous work, work a human will review later, or /ds-ledger.
In this repository28 other files it gives its agents.
- ds-adoptSkill
- ds-best-of-nSkill
- ds-blast-radiusSkill
- ds-blueprintSkill
- ds-catchupSkill
- ds-context-resetSkill
- ds-design-auditSkill
- ds-designSkill
- ds-explainSkill
- ds-helpSkill
- ds-jurySkill
- ds-lessonsSkill
- ds-modeSkill
- ds-plan-loopSkill
- ds-pr-commentsSkill
- ds-principlesSkill
- ds-prSkill
- ds-readySkill
- ds-review-loopSkill
- ds-reviveSkill
- ds-sharpenSkill
- ds-shipSkill
- ds-swarmSkill
- ds-tddSkill
- ds-unslopSkill
- ds-unstickSkill
- ds-verify-loopSkill
- ds-watch-prSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 7 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
Create one TSV file, for example .ds-ledger.tsv, with the header ts phase decision why evidence result.
- Distinctive: line 13
Before each step, re-read the goal row and the last ten rows.
- Distinctive: line 14
After each real decision, append one row:
- Distinctive: line 21
Keep each cell on one line with no tabs.
- Distinctive: line 28
If three rows in a row show no progress toward the goal, stop the run and report.
- Distinctive: line 32
Every row needs evidence you can open or rerun.
- Distinctive: line 33
Write rows the way you would tell a teammate, plain and concrete.
- Distinctive: line 34
Do not rewrite old rows. Append corrections.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
ds-ledger
A long run forgets its own reasons. A short written ledger keeps the goal in view and lets a reviewer trust the result.
Steps
- Create one TSV file, for example
.ds-ledger.tsv, with the headerts phase decision why evidence result. Put the goal and done condition in the first row. - Before each step, re-read the goal row and the last ten rows. Check that the next step still serves the goal. If it does not, log the drift and correct course.
- After each real decision, append one row:
- ts: ISO 8601 time.
- phase: the phase or workstream.
- decision: what you chose or did, one line.
- why: the reason in plain words.
- evidence: a pointer, never prose. A commit hash,
file:line, a command and its result, or an artifact path. - result: the outcome, such as "tests pass", "reverted", "open".
- Keep each cell on one line with no tabs.
- Log reverts and dropped work too, with the reason. Failures are the most useful rows.
- At the end, add a summary row and point the user to the file. Commit it only if a reviewer needs it.
Stop when
- The run ends and the summary row is written.
- If three rows in a row show no progress toward the goal, stop the run and report.
Rules
- Every row needs evidence you can open or rerun. "Looked fine" is not evidence.
- Write rows the way you would tell a teammate, plain and concrete.
- Do not rewrite old rows. Append corrections.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

