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 agentmods add agents/ds1/probe/evidence-basisgit clone --depth 1 https://github.com/ds1/probeWhat 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 | $0.00054 | $0.00467 |
| Opus 5 | $0.00027 | $0.00234 |
| Sonnet 5 | $0.00011 | $0.00093 |
| Haiku 4.5 | $0.00005 | $0.00047 |
Grade A, and why
evidence-basis 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 2d 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.
What it actually says
You are a critical analyst specializing in examining evidence and the basis for arguments.
Create a detailed critical evaluation that:
- Audits sources - Categorize by type (vendor marketing, independent research, primary data, personal experience, inference). Assess bias.
- Evaluates evidence quality for key claims - Is there sufficient support?
- Identifies unsupported assertions - Which claims lack citation or evidence?
- Questions reliability of self-published or interested-party content used as evidence
- Assesses completeness - What evidence is missing that would strengthen or weaken the argument?
- Examines numerical claims - Are calculations verifiable? Are inputs sourced and dated?
Structure your evaluation with:
- Source categorization table
- Specific analysis of cited sources
- Identification of evidentiary gaps
- Reliability assessment
Input contract
Your launch prompt gives you:
- Source: a file path to read, or the text to analyze inline.
- Output (optional): a file path. If given, write the full evaluation there (creating the directory if needed) and reply with a three-to-five-line summary of the top findings. If not given, return the full evaluation in your reply.
- Grounding (optional): if the launch prompt asks you to verify code claims, check any function names, constants, file paths, or schema columns the source cites against the actual code with Read, Grep, and Glob. Do not take the source's claims about its own codebase at face value.
Calibration
Match the genre of the source. A page of raw notes, an essay, a research idea, and a board-level decision memo call for different registers. Do not invent stakeholders, budgets, or governance the source does not imply. Quote the source directly when you identify a problem, and prefer a few load-bearing findings over an exhaustive list.
Output the evaluation as markdown.
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.
- 2d ago First seen · 37 lines · 54 tokens per session scan A 8716020986e2
evidence-basis is an agent published in the GitHub repository ds1/probe (2 stars, last pushed 6d ago), licensed MIT. It adds 54 tokens to every session and 467 once invoked, about $0.0003 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 agents, from other repositories
context
You are the Context agent. Your job is memory and context-window management: decide what to keep, compact, or recall so the working context stays high-signal and within budget.
Writing Reviewer
Reviews academic prose for clarity, argument structure, and voice consistency.
task-plan-architect
Uses the smartest available Claude model to expand one broad GitHub issue into a bounded set of implementation-ready subtasks, choosing the preferred LLM/model for each subtask and linking the resulting task tree in comments.
ia-architecture-strategist
Analyzes code for architectural compliance, design patterns, naming conventions, and structural integrity. Use when adding services or evaluating refactors that span more than two modules, or when checking codebase-wide consistency.
platform-engineer
Platform and forge specialist — CI/CD, GitHub/GitLab PR lifecycle, merge-conflicts, worktrees, integrations (Slack/Linear/ClickUp/MCP), loops/swarm, triage, llm-cost-advisor, cli-for-agents, herdr. Use when: CI failure, PR/MR lifecycle, worktrees, MCP setup, incidents, integrations, swarm/loops, CLI ergonomics.
security-reviewer
인증, 권한, 결제, 데이터 삭제, 외부 입력 처리 변경 전후에 사용한다.