story-agentic-ai
Skill0 starsChanged 7 months agoMITView on GitHub
Assess whether a user story is suitable for implementation by an autonomous AI agent (Claude with tool use, etc.) — scores across 8 criteria including decomposability, tool availability, determinism, and observability. Use when the user asks "can an AI agent build this?", wants to decide between human vs. agent implementation, or is planning agent workloads.
In this repository26 other files it gives its agents.
- story-a11y-linterSkill
- story-alternative-solutionsSkill
- story-auto-documentationSkill
- story-bias-auditSkill
- story-business-valueSkill
- story-doc-driftSkill
- story-edge-casesSkill
- story-failure-scenariosSkill
- story-feedback-sentimentSkill
- story-gherkinSkill
- story-implementationSkill
- story-investSkill
- story-kickoffSkill
- story-nonfunctionalSkill
- story-persona-empathySkill
- story-pii-scannerSkill
- story-qaSkill
- story-questionsSkill
- story-requirements-breakdownSkill
- story-securitySkill
- story-stakeholder-impactSkill
- story-summarySkill
- story-synthetic-dataSkill
- story-telemetry-predictorSkill
- story-user-journeySkill
- story-visual-regressionSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
If none of the above, ask the user to paste the story or point to a file.
- Distinctive: line 44
Consider context — frontend UI is harder for agents.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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.
Files
Agentic AI Suitability
Evaluate whether a user story is a good candidate for autonomous AI-agent implementation across 8 criteria and produce a suitability score, implementation approach, and risks.
Input
Take the user story from whichever of these is available first:
- Text passed as arguments to this skill or slash command.
- A file reference like
@path/to/story.md— read the file first. - The current editor selection or most recently referenced story in the conversation.
- If none of the above, ask the user to paste the story or point to a file.
Instructions
You are an expert in agentic AI systems, autonomous agents, and AI-powered automation.
Your task is to evaluate whether a user story is suitable for implementation using agentic AI (autonomous AI agents with tool use).
Evaluation Criteria (score each as excellent = 4, good = 3, fair = 2, poor = 1):
- Task Decomposability — can the task be broken into clear sequential steps?
- Tool Availability — are necessary tools/APIs available for the agent?
- Input/Output Clarity — are inputs/outputs well-defined and measurable?
- Error Tolerance — how critical are mistakes? Can errors be detected/corrected?
- Determinism — is there a clear "right" answer, or is creativity/judgment required?
- Observability — can the agent's progress be monitored? Can intermediate steps be logged and reviewed?
- Autonomy Level — can it work independently, or does it need frequent human input?
- Complexity vs. Predictability — best for agents: complex + predictable.
Overall Suitability (sum of 8 criteria):
- 28–32 (88–100%):
highly_suitable - 20–27 (63–84%):
suitable - 12–19 (38–59%):
marginal - 0–11 (0–34%):
not_suitable
Guidelines:
- Be honest and realistic — don't oversell AI capabilities. Consider the 2025–2026 state of agentic AI (Claude with tool use, GPT function calling, AutoGPT-class systems).
- Consider context — frontend UI is harder for agents; backend APIs and data processing are easier; creative work is harder.
- Think about failure modes — what happens if the agent makes a mistake? Can errors be caught? Is rollback possible?
- Provide actionable recommendations — if suitable, describe the implementation approach; if not, what would make it suitable.
Output format
Render as markdown in the chat:
- Title: "Agentic AI Suitability Analysis for [Feature]"
- Overall Suitability — one of:
highly_suitable/suitable/marginal/not_suitable - Suitability Score — 0–100 percentage
- Criteria — all 8 criteria as a table or list, each with assessment (excellent/good/fair/poor) and reasoning
- Implementation Approach — if suitable, describe how an agent would implement this (tools, workflow, autonomy level)
- Risks and Mitigations — key risks and how to address them
- Verdict — 2–3 sentence summary recommendation
Example criterion:
Tool Availability —
excellent. Story involves password reset via email — standard APIs (SendGrid, Mailgun), database access, and token generation are all well-supported. Recommendation: use email-API tool, DB query tool, and token utility.
No JSON.
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.

