How do I install the Analyze skill?

Run npx skillmds@latest add telum-ai/analyze in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Analyze safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Analyze?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Analyze free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Analyze?

telum-ai (@telum-ai) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Analyze by telum-ai — Agent Skill

Name: Analyze
Rating: 4.666666666666667 (3 reviews)
Author: telum-ai

File contents

analyze

Canonical planning-analysis engine. $ARGUMENTS may contain --level project|epic|story.

Select level and depth

Honor explicit --level; else infer story, epic, or project from the target path.
Project requires project.md, PRD.md, epics.md; epic requires epic.md, epic-tech-spec.md, epic-breakdown.md; story requires spec.md, plan.md, tasks.md. STOP on a missing prerequisite.
Read .speck/project.json (play_level; missing = Platform). Count epics from epics.md and epics/ directories.
Story: Sprint skips; Build runs S1; Platform runs S1-S3. Project: Build 1-3 runs focused L7 only when requested, Build 4+ runs L3/L6/L7, Platform runs L1-L7. Epic: Build runs focused L7 and Platform runs L1-L7.

Load the core

Before analysis, run one receipted load. It supplies scope, flow-fit, promise, traceability, severity, and role-separation rules without loading any reviewer lens or report template:

python3 .speck/scripts/context/speck_context.py analyze-core \
  --select level=<project|epic|story>

Require exit 0 and SPECK_CONTEXT_RECEIPT. Report templates and references/reports/* are forbidden until findings return.

Dispatch lenses

For every required lens id, dispatch one reviewer that did not author the corpus. That reviewer runs exactly one level-specific loader and receives the target artifact list:

# project or epic
python3 .speck/scripts/context/speck_context.py analyze-<project|epic>-lens \
  --select lens=L#

# story
python3 .speck/scripts/context/speck_context.py analyze-story-lens \
  --select lens=S#

The loader emits references/lens-spine.md plus exactly one of references/lenses/project/L#.md or references/lenses/epic/L#.md; sibling level and lens nodes are forbidden. The conductor does not preload lens nodes. Lenses do not share findings before verification.

Finish

After findings return, run exactly one second-stage load before any report mutation:

python3 .speck/scripts/context/speck_context.py analyze-report \
  --select level=<project|epic|story>

It emits the selected template plus exactly one level report contract. Follow it to verify findings, write the report, run gates, commit after the analyzed corpus, and declare BLOCKED | NEEDS_FIXES | CLEAN. Read references/gate-codes.md only when enforcing or explaining analysis codes. STOP on any loaded-node STOP.

---name: analyzedescription: Runs decorrelated planning analysis. Use after project-plan, epic-breakdown, or story-tasks before downstream work.---# analyzeCanonical planning-analysis engine. `$ARGUMENTS` may contain `--level project|epic|story`.## Select level and depth1. Honor explicit `--level`; else infer story, epic, or project from the target path.2. Project requires `project.md`, `PRD.md`, `epics.md`; epic requires `epic.md`, `epic-tech-spec.md`, `epic-breakdown.md`; story requires `spec.md`, `plan.md`, `tasks.md`. STOP on a missing prerequisite.3. Read `.speck/project.json` (`play_level`; missing = Platform). Count epics from `epics.md` and `epics/` directories.4. Story: Sprint skips; Build runs S1; Platform runs S1-S3. Project: Build 1-3 runs focused L7 only when requested, Build 4+ runs L3/L6/L7, Platform runs L1-L7. Epic: Build runs focused L7 and Platform runs L1-L7.## Load the coreBefore analysis, run one receipted load. It supplies scope, flow-fit, promise, traceability, severity, and role-separation rules without loading any reviewer lens or report template:```bashpython3 .speck/scripts/context/speck_context.py analyze-core \ --select level=<project|epic|story>```Require exit 0 and `SPECK_CONTEXT_RECEIPT`. Report templates and `references/reports/*` are forbidden until findings return.## Dispatch lensesFor every required lens id, dispatch one reviewer that did not author the corpus. That reviewer runs exactly one level-specific loader and receives the target artifact list:```bash# project or epicpython3 .speck/scripts/context/speck_context.py analyze-<project|epic>-lens \ --select lens=L## storypython3 .speck/scripts/context/speck_context.py analyze-story-lens \ --select lens=S#```The loader emits `references/lens-spine.md` plus exactly one of `references/lenses/project/L#.md` or `references/lenses/epic/L#.md`; sibling level and lens nodes are forbidden. The conductor does not preload lens nodes. Lenses do not share findings before verification.## FinishAfter findings return, run exactly one second-stage load before any report mutation:```bashpython3 .speck/scripts/context/speck_context.py analyze-report \ --select level=<project|epic|story>```It emits the selected template plus exactly one level report contract. Follow it to verify findings, write the report, run gates, commit after the analyzed corpus, and declare `BLOCKED | NEEDS_FIXES | CLEAN`. Read `references/gate-codes.md` only when enforcing or explaining analysis codes. STOP on any loaded-node STOP.

Frequently asked questions

Analyze