How do I install the Story Validate skill?

Run npx skillmds@latest add telum-ai/story-validate 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.

What does the Story Validate skill do?

Validates a story after audit and declares readiness. Use at the story prove gate before retrospective. It is listed under Security on SkillMD.

Is Story Validate 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 Story Validate?

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 Story Validate free to use?

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

Who published Story Validate?

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

Story Validate by telum-ai — Agent Skill

Name: Story Validate
Rating: 4.2 (5 reviews)
Author: telum-ai

File contents

story-validate

Prereq: spec.md, plan.md, tasks.md, audit-report.md. Verdict = readiness state (never PASS/FAIL). Output: [STORY_DIR]/validation-report.md.

Cheap-key read only: locate STORY_DIR; read .speck/project.json; take claimed_state from --claim, the requested target, or the existing report claim. Classify UI vs backend.
Before the first mutation, run exactly one contract:
- Backend: python3 .speck/scripts/context/speck_context.py story-validate-backend --select claimed_state=<state>
- UI: python3 .speck/scripts/context/speck_context.py story-validate-ui --select claimed_state=<state>
Require exit 0 and SPECK_CONTEXT_RECEIPT. Do not directly load unselected state, archetype, or axis references. The receipt's post_write_gates, post_write_gates_all, and gate_policy fields are the exact executable closure contract; do not reconstruct or batch it.
Execute every receipted node in order; record build_sha, cross-examine the audit, LARP, visual, and runtime evidence already produced, run only the validation probes, and lower the claim to what is earned. Follow the receipted post-write node literally: after the latest stamp, run every gate individually even on a downgrade. Correctable red may edit, re-stamp, and rerun all; otherwise run the remaining gates, record conformant_red blockers, then make no edit.

STOP on any node STOP / P0 / missing required evidence.

Frequently asked questions

Story Validate