agentleFS
synaptiaisynaptiai/bdsk/skills/assume/SKILL.md

assume

Skill0 starsChanged 7 months agoMITView on GitHub

Capture an assumption as a structured BDSK assumption_record artifact. Use when a decision, belief, or unresolved inference affects implementation.

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 2 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Clarify the assumption: If the statement is vague, ask for specifics.

    line 25
  2. Distinctive:

    Assess impact: Ask the user or infer:

    line 27
  3. Very specific:

    Do NOT use assumption: with statement/source/impact — use spec: with all 7 required fields (statement, rationale, source_type, source_refs, …

    line 71
  4. Very specific:

    Do NOT use bare string IDs in trace.downstream — use {target_id: <id>, edge: <edge>} objects

    line 72
  5. Very specific:

    Do NOT use approver/date/decision in approvals — use authority_role/approver/approved_at

    line 73
  6. Very specific:

    Do NOT use depends_on upstream from assumption_record (not in compatibility matrix)

    line 81

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/assume

Capture an assumption as a structured record.

Input

The user provides a statement: $ARGUMENTS

Process

  1. Clarify the assumption. If the statement is vague, ask for specifics. An assumption must be a concrete claim that could be true or false.
  2. Assess impact. Ask the user or infer:
    • Impact level: low, medium, high, or critical
    • Area: security, product, architecture, compliance, or other
  3. Write the artifact:
kind: assumption_record
schema_version: "0.3"
id: AR-<short-slug>-<timestamp>
title: <short assumption title>
status: proposed
owners:
  - <current user>
created_at: <current ISO-8601 timestamp>
updated_at: <current ISO-8601 timestamp>
trace:
  upstream: []
  downstream: []
approvals: []
metadata:
  impact_level: <low|medium|high|critical>
  area: <security|product|architecture|compliance|other>
spec:
  statement: <the assumption itself, precise and falsifiable>
  rationale: <why this assumption exists>
  source_type: <documented|inferred|decision|external-standard|legacy-system>
  source_refs: []
  decision_authority: <role or person who can accept/reject this>
  review_by: <ISO-8601 date, default 30 days from now>
  resolution_rule: <what must happen before execution if unresolved>
  1. Write to disk. Save to artifacts/assumptions/<id>.yaml
  2. Present for review. The user runs /approve <id> to accept the assumption.

Rules

  • Assumptions MUST be falsifiable statements, not vague beliefs
  • Impact level MUST reflect the consequence of the assumption being wrong
  • review_by defaults to 30 days from creation unless the user specifies otherwise
  • resolution_rule MUST describe what blocks execution if this assumption is unresolved
  • Schema compliance: The artifact MUST use the exact template structure above. Common mistakes to avoid:
    • Do NOT use assumption: with statement/source/impact — use spec: with all 7 required fields (statement, rationale, source_type, source_refs, decision_authority, review_by, resolution_rule)
    • Do NOT use bare string IDs in trace.downstream — use {target_id: <id>, edge: <edge>} objects
    • Do NOT use approver/date/decision in approvals — use authority_role/approver/approved_at
    • metadata MUST contain both impact_level and area — these are required fields
    • Valid authority roles (5 total): product_authority, technical_authority, security_authority, release_authority, qa_authority. Do NOT use spec_approver, plan_approver, or invented role names.
    • Valid trace edges (10 total): depends_on, derived_from, constrains, implements, proves, evaluates, produced_by, supersedes, escalates_to, waives.

Edge-kind rules for assumption_record

  • Upstream derived_from → behavior_spec, assumption_record, contract_artifact
  • Downstream constrains → behavior_spec, generated_diff
  • Do NOT use depends_on upstream from assumption_record (not in compatibility matrix)

Download now (ZIP)

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. MarcosCamara01/sddguard★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository11 other files it gives its agents.

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

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.