gaps
Skill1 starsChanged 8 days agoMITView on GitHub
Use when the builder needs to identify missing information, unresolved questions, or misalignments with the business team. Also use when preparing questions for the next meeting, or when the coding agent hits a blocker due to missing business context.
5 other projects use a version of this.
Adopted by
Most recently taken up by
- ansvisor/ansvisortheir own version, first seen Nov 8, 2026
- ai-search-guru/getcito-worlds-first-open-source-aio-aeo-or-geo-tooltheir own version, first seen Nov 8, 2026
- yuri-xyz/skillstheir own version, first seen Nov 8, 2026
- Consortium-team/project-creatortheir own version, first seen Sep 10, 2026
- anthropics/claude-for-legaltheir own version, first seen May 15, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- anthropics/claude-for-legal★ 9.5ksame name
- ai-search-guru/getcito-worlds-first-open-source-aio-aeo-or-geo-tool★ 432same name
- ansvisor/ansvisor★ 233same name
- Consortium-team/project-creator★ 17same name
- yuri-xyz/skills★ 2same 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.
The file
Show its rules and text
We split this skill into the rules it gives an agent
17 rules. 0 are common, 10 are distinctive, and 7 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
10
Distinctive
This author's own choices. Worth a look.
7
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
When this skill is activated, your FIRST line of output MUST be:
- Very specific: line 29
After digest — automatically review new knowledge for gaps
- Very specific: line 30
Before prd — ensure we have enough to write a solid spec
- Very specific: line 31
When work hits a blocker — coding agent can't proceed without clarity
- Very specific: line 32
Before meetings — generate question list (prepare-meeting calls this)
- Distinctive: line 39
Read in this order:
- Distinctive: line 48
Look for these categories:
- Very specific: line 63
Create or update: docs/specs/<epic>/gaps.md (or docs/knowledge/gaps.md if no epic yet)
- Very specific: line 105
If applicable: "I can prepare a meeting briefing — run prepare-meeting"
- Distinctive: line 109
When the builder provides answers (after a meeting, chat, etc.):
- Very specific: line 111
Update the gap status to resolved
- Distinctive: line 113
Update the relevant knowledge base document
- Distinctive: line 114
If the answer changes the PRD → flag: "This answer affects [story X].
- Distinctive: line 115
If the answer unblocks kanban tasks → flag: "T-004 is now unblocked."
- Distinctive: line 126
Every gap MUST have an "Impacts" field.
- Distinctive: line 128
Never close a gap without recording the answer.
- Distinctive: line 130
If a gap has been open for multiple sessions, escalate its priority and suggest action.
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
OMT Gaps — Information Gap Analysis & Question Tracking
You are the quality control brain of the One Man Team. Your job is to find what's missing — the questions nobody asked, the assumptions nobody verified, the edge cases nobody considered — and track them until they're resolved.
Language
Detect the user's language from their first message and respond in the same language throughout.
Activation Announcement
When this skill is activated, your FIRST line of output MUST be:
[OMT/gaps] <brief description of what you're about to do>Examples:
[OMT/gaps] Starting...
This helps the builder always know which skill is driving the current response. If you transition to a different skill mid-conversation, announce the switch.
When to Use
- After
digest— automatically review new knowledge for gaps - Before
prd— ensure we have enough to write a solid spec - When
workhits a blocker — coding agent can't proceed without clarity - Before meetings — generate question list (
prepare-meetingcalls this) - Builder asks: "what don't we know?" or "what should I ask them?"
Procedure
Step 1: Scan Sources
Read in this order:
docs/knowledge/— what we knowdocs/specs/*/prd.md— what we're trying to builddocs/timeline/— what was discussed (look for open questions)docs/specs/*/kanban.md— what's blocked- Existing
docs/specs/*/gaps.md— what we already identified
Step 2: Gap Detection
Look for these categories:
| Category | What to look for |
| Undefined behavior | User stories without edge case coverage. "What happens when X fails?" |
| Missing data | Fields, formats, or sources mentioned but not specified |
| Conflicting information | Two sources say different things |
| Unvalidated assumptions | Things the BA assumed but nobody confirmed |
| Missing stakeholder input | Decisions that need sign-off |
| Technical unknowns | Integration points with no documentation |
| Scope ambiguity | Features described vaguely — could mean multiple things |
| Dependency gaps | External systems/teams we need but haven't confirmed |
Step 3: Produce/Update Gaps Document
Create or update: docs/specs/<epic>/gaps.md (or docs/knowledge/gaps.md if no epic yet)
Format:
# Gaps & Open Questions: <Epic/Project Name>
**Last Updated:** YYYY-MM-DD
**Total Gaps:** [N] open, [N] resolved
## Critical (Blocks Design/Implementation)
| ID | Question | Category | Impacts | Source | Ask Who | Status |
|----|----------|----------|---------|--------|---------|--------|
| G-001 | [specific question] | [category] | [what's blocked] | [timeline ref] | [person/role] | open |
| G-002 | ... | ... | ... | ... | ... | open |
## Important (Needed Before Milestone Completion)
| ID | Question | Category | Impacts | Source | Ask Who | Status |
|----|----------|----------|---------|--------|---------|--------|
| G-010 | ... | ... | ... | ... | ... | open |
## Nice to Know (Clarification, Not Blocking)
| ID | Question | Category | Impacts | Source | Ask Who | Status |
|----|----------|----------|---------|--------|---------|--------|
| G-020 | ... | ... | ... | ... | ... | open |
## Resolved
| ID | Question | Answer | Answered By | Date | Updated In |
|----|----------|--------|-------------|------|------------|
| G-003 | [question] | [answer] | [person] | YYYY-MM-DD | [which doc was updated] |Step 4: Report to Builder
Summarize:
- "[N] new gaps found, [N] total open"
- "Critical blockers: [list the most important 1-3]"
- "Recommended: bring these to your next meeting with [stakeholder]"
- If applicable: "I can prepare a meeting briefing — run
prepare-meeting"
Resolving Gaps
When the builder provides answers (after a meeting, chat, etc.):
- Update the gap status to
resolved - Record the answer, who provided it, and the date
- Update the relevant knowledge base document
- If the answer changes the PRD → flag: "This answer affects [story X]. PRD should be updated."
- If the answer unblocks kanban tasks → flag: "T-004 is now unblocked."
Auto-Detection During Other Skills
Other OMT skills should call gap detection logic:
omt-digest: after processing, check for new gapsomt-prd: during generation, flag assumptions as gapsomt-work: when hitting an unknown, create a gap entry
Rules
- Every gap MUST have an "Impacts" field. Gaps without impact are noise.
- Prioritize ruthlessly. A builder's time with stakeholders is limited — surface what matters most.
- Never close a gap without recording the answer. The answer becomes knowledge.
- Frame questions clearly enough that a non-technical person can answer them.
- If a gap has been open for multiple sessions, escalate its priority and suggest action.
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.

