Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Abhillashjadhav/AI-PM-essential-skills --skill ai-feature-kill-criteriagit clone --depth 1 https://github.com/Abhillashjadhav/AI-PM-essential-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/abhillashjadhav/ai-pm-essential-skills/ai-feature-kill-criteria)<a href="https://agentmods.dev/skills/abhillashjadhav/ai-pm-essential-skills/ai-feature-kill-criteria"><img src="https://agentmods.dev/badge/skills/abhillashjadhav/ai-pm-essential-skills/ai-feature-kill-criteria/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/abhillashjadhav/ai-pm-essential-skills/ai-feature-kill-criteria"><img src="https://agentmods.dev/badge/skills/abhillashjadhav/ai-pm-essential-skills/ai-feature-kill-criteria.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00072 | $0.00897 |
| Opus 5 | $0.00036 | $0.00449 |
| Sonnet 5 | $0.00014 | $0.00179 |
| Haiku 4.5 | $0.00007 | $0.00090 |
Grade A, and why
ai-feature-kill-criteria scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Feature Kill Criteria
Prevent attractive AI demos from turning into open-ended investments.
Required input
Ask for only what is missing:
- The user problem and target user.
- The proposed AI behavior.
- The business or customer outcome expected.
- Known constraints such as latency, cost, privacy, safety, or workflow fit.
- The maximum evidence budget: time, users, data, or engineering effort.
Do not accept "we will know when we see it" as a success definition.
Workflow
1. Separate the demo from the product claim
Write the product claim in this form:
For [user], [AI behavior] will improve [observable outcome] because [mechanism].
Flag any part that is an aspiration rather than evidence.
2. Extract the assumptions
Create 3-7 assumptions across these categories:
- problem: the pain is frequent and important;
- behavior: users will trust, understand, and act on the output;
- capability: the system can perform on realistic inputs;
- workflow: the feature fits the real sequence of work;
- economics: latency and cost are viable at expected volume;
- risk: privacy, safety, compliance, and misuse remain inside acceptable bounds;
- adoption: the target user can discover and repeatedly use it.
Each assumption must be falsifiable. Replace vague wording such as "users will like it" with a measurable observation.
3. Define a kill threshold for every critical assumption
Use this format:
| Assumption | Evidence to collect | Continue threshold | Kill threshold | Decision owner |
|---|
A kill threshold must be observable and time-bounded. Examples:
- fewer than 4 of 10 target users complete the task without expert help;
- factual accuracy remains below 95% on the original failure set after two model or prompt iterations;
- p95 latency exceeds 3 seconds under expected concurrency;
- cost per completed task exceeds the current human-assisted path;
- more than 2% of high-risk outputs require correction;
- the workflow saves less than five minutes per task after two weeks of use.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 113 lines · 72 tokens per session scan A 582c18848bd0
ai-feature-kill-criteria is a skill published in the GitHub repository Abhillashjadhav/AI-PM-essential-skills (4 stars, last pushed 10d ago), licensed MIT. It adds 72 tokens to every session and 897 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
implementing-llm-guardrails-for-security
Implements input and output validation guardrails for LLM-powered applications to prevent prompt injection, data leakage, toxic content generation, and hallucinated outputs. Builds a security validation pipeline using NVIDIA NeMo Guardrails Colang definitions, custom Python validators for PII detection and content…
vs-search-tuning
Use when a user asks an agent to evaluate or tune text search similarity for an existing Viking AI Search application and dataset.
regex-vs-llm-structured-text
Decision framework for choosing between regex and LLM when parsing structured text — start with regex, add LLM only for low-confidence edge cases.
rag-construction
Build RAG systems for construction knowledge bases. Create searchable AI-powered construction document systems.
ml-model-builder
Build ML models for construction predictions. Train and evaluate custom models for cost, duration, and risk prediction.
schedule-forecaster
Predict project completion dates using ML models. Forecast schedule delays based on current progress, historical patterns, and risk factors.