traceable-decision-log
/traceable-decision-log
Use when writing or auditing a decision sheet or status brief, so that each open item carries verifiable provenance, a stop-on-contradiction control, and a timestamp of its data.
- Author
- otixe ↗
- Repository
- agent-skills ↗
- Licence
- MIT
- Last repository change
- 26.10.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 26.10.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Use when writing or auditing a decision sheet or status brief, so that each open item carries verifiable provenance, a stop-on-contradiction control, and a timestamp of its data.
Original author description
Use when writing or auditing a decision sheet or status brief, so that each open item carries verifiable provenance, a stop-on-contradiction control, and a timestamp of its data.
How it works
- Authorization and test boundary
- Step 0: Admission test for any control
- Step 1: Provenance card per item
- Step 2: Check contradictions in both forms
- Step 3: Blocking claims need the last event
- Step 4: Timestamp the data
- Step 5: Do not publish or send without re-checking
- Controls
- Done when
SKILL.md preview
--- name: traceable-decision-log description: "Use when writing or auditing a decision sheet or status brief, so that each open item carries verifiable provenance, a stop-on-contradiction control, and a timestamp of its data." --- # Traceable Decision Log ## Authorization and test boundary These instructions grant no permission. Confirm the exact resources, allowed effects and data handling before acting. Use synthetic canaries and disposable fixtures for deliberate failures; never use real credentials as test inputs. Keep secrets and fragments out of model context, arguments, logs and plaintext artifacts. Imports, constructors, help and dry-run commands may have effects. Establish external-effect isolation before executing untrusted code; application stubs are not a sandbox. Record untested scope and obtain separate authorization for production changes or destructive actions. Use this before escalating a decision to a human, or when writing or auditing any document that consolidates pending decisions, such as a decision sheet or status brief. The goal is that a contradiction hidden across many sections becomes a visible empty field. ## Step 0: Admission test for any control Before adopting a control, run it against the incident that motivated it. If it would pass that incident without catching it, discard it or redesign it. A control that shows green on the exact failure it was built for is worse than none, because it adds an appearance of diligence. Also list every store where authority can live, such as policy tables, task comments, activity logs, and memory notes. A control that queries one of four stores will miss the decision most of the time. ## Step 1: Provenance card per item Each decision item carries three search-derived fields: - authority_checked: store, identifier, last modified time, and query time. - last_decision_found: date, author, and a verbatim quote, or NONE. - why_still_open: the reason the item remains open despite the last decision. Rule: if last_decision_found is not empty, the item cannot be labeled pending unless why_still_open is filled. Citing a policy without its identifier and modification time is decoration, and it discourages verification. ## Step 2: Check contradictions in both forms A recommendation can contradict a decision; a presupposition can contradict it too. A question such as "which of the two is correct?" may assume that both are correct. Verify the premises of your questions, not only your recommendations. ## Step 3: Blocking claims need the last event Do not write that an item blocks another without quoting the latest event of the blocked item, with its timestamp. If that event is an unblock, completion, or a scope reduction, the claim is invalid. ## Step 4: Timestamp the data A decision sheet is a snapshot. State when its data was verified and when it was delivered. Date a document by its content: the newest event it reports as past is a floor, and the oldest event it reports as not occurred but that did occur is a ceiling. ## Step 5: Do not publish or send without re-checking Items that touch external publication are irreversible. Re-verify them immediately before any send. ## Controls - Negative: no item claims pending while a last decision exists without a reason. - Positive: the control flags a seeded contradiction in a test document. ## Done when Every item has the three provenance fields, timestamps are present, and contradictions are listed for the owner to resolve.
Versions
The database has no other skills named traceable-decision-log.
Database snapshot: 13.11.2026 · package version c6adc12322d7 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 86log-decision
Record a decision or an accepted risk in the decision log. Use when the owner or the team makes a decision. - 100brief
The front door to PM Pilot. Turns one decision question into an evidence-backed decision brief where every claim carries its source, counter-evidence is stated, and missing evidence is named rather than papered over. Works on a local folder with no MCP setup. Triggers on:… - 90create-decision-log
Record a product or engineering decision with full context when the user asks to log a decision, record a choice, or document why something was decided - 90decision-log
Creates a structured decision log from project changes, scope revisions, or PM decisions. Use whenever a decision needs to be formally recorded - including when someone says "log this decision", "document this change", "we changed direction on X", or when the PM Orchestrator… - 100doncheli-audit-trail
Record and query the decision log for a project. Activate when user mentions "audit", "trail", "log decisions", "decision history", "why was this decided", "ADR", "architecture decision". - 100decision-log
Extracts structured decisions from meeting notes, Slack threads, or email chains and maintains a decision log separate from action items. Each decision is captured with context, alternatives considered, participants, date, and related tasks or risks. Supports two modes: new log…