writing-specs
Skill249 starsChanged 15 days agoMITView on GitHub
Use when agent_specs is on (or the user asks for a spec) to write what done looks like and why before a plan or implementation.
8 other projects use a version of this.
Adopted by
Most recently taken up by
- victorhqc/filberttheir own version, first seen Nov 4, 2026
- JetBrains/thinkrailtheir own version, first seen Oct 13, 2026
- whiteguo233/OpenBiliClawtheir own version, first seen Oct 8, 2026
- sergeyklay/.agentstheir own version, first seen Sep 28, 2026
- fireshort/skillstheir own version, first seen Aug 26, 2026
- augments-labs/sdlc-skillstheir own version, first seen Aug 8, 2026
Used with
Projects with it also use writing-plans.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
OpenAISkillwriting-plans
Use when you have a spec or requirements for a multi-step task, before touching code
Also in 3 projects that use this
Other versions8 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 8
- JetBrains/thinkrail★ 7.9ksame name
- whiteguo233/OpenBiliClaw★ 3.7ksame name
- devstefancho/skills★ 13same name
- sergeyklay/.agents★ 6same name
- victorhqc/filbert★ 4same name
- augments-labs/sdlc-skills★ 1same name
- fireshort/skills★ 1same name
- xmtplabs/code-factory★ 1same 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 repository15 other files it gives its agents.
- AGENTS.mdAGENTS.md
- commands/agents.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .cursor/rules/feature-worktrees.mdcCursor rule
- .cursor/rules/no-cursor-commit-notes.mdcCursor rule
- diagnosisSkill
- draftingSkill
- feature-prototypeSkill
- inventingSkill
- mvp-prototypeSkill
- proofSkill
- shippingSkill
- swarmSkill
- testingSkill
- using-superliteSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 0 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 8 edited versions we read keep these.
0
Distinctive
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
- Very specific: line 12
Check bash "$CLI" agent_specs status.
- Very specific: line 13
Feature flow (default): write one feature-sized spec before implementers code (and before or beside the .feature.md / machine plan as the …
- Very specific: line 14
No human gate on task-local notes.
- Very specific: line 27
When status shows open_in_harness: on, after writing the file, open it for the user (superlite:drafting, After writing.
“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Writing specs
A spec defines what “done” looks like and why. A plan chooses how to get there. One spec can outlive many plans: rewrite or replace the plan without rewriting success criteria. The spec is durable authority; if a plan drifts, fix the plan or report DONE_WITH_CONCERNS—never silently override the spec.
When
- Check
bash "$CLI" agent_specs status. If off, skip unless the user asked for a spec. - Feature flow (default): write one feature-sized spec before implementers code (and before or beside the
.feature.md/ machine plan as the success criteria). Optional task-local notes under.superlite/specs/<plan-topic>/refine how for a long-running task; they must stay consistent with the feature spec. - No human gate on task-local notes. Feature-level specs and plans may be opened for the user when
open_in_harnessis on.
Paths
- Feature (or sub-feature) spec:
docs/plans/YYYY-MM-DD-<topic>.spec.md(or the path the user names). - Task-local note (optional):
.superlite/specs/<plan-topic>/T<N>-<slug>.md.
Shape
Use templates/feature-spec.md. Prefer diagrams and tables; keep prose outside fences and tables ≤ 150 words for task-local notes. Feature specs may be longer but stay dense: intent/why, goals and non-goals, constraints, what “done” looks like (acceptance), decided vs left to the builder. Mermaid first for structure, flow, or state.
Open in harness
When status shows open_in_harness: on, after writing the file, open it for the user (superlite:drafting, After writing; the harness's open row is in using-superlite/references/<harness>-tools.md). The flag now covers plans and specs. superlite-cli agent_specs open off turns it off without disabling specs.
Style (flows §9)
| Rule | Instead of |
| Diagram first for structure, flow, or state | Paragraphs of boxes and arrows |
| Tables for contracts, options, edge cases | Long bullet prose |
| Precise technical register | Vague paraphrase |
| Every sentence a fact, constraint, or decision | Filler and restatement |
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.