codesage-finding-verifier

An agent that challenges up to ten newly reported code problems from one feature area. It reads the cited code and returns whether each claim is confirmed, refuted, or uncertain.

In plain words
What is it for?
Verify reported bugs, security issues, performance problems, or maintenance concerns by tracing callers, contracts, downstream handling, and tests.
Why use it?
It adds a second check before findings are accepted, so unsupported claims are separated from reachable defects.

Agent

Install

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.

agentmods
npx agentmods add agents/iliaal/codesage/codesage-finding-verifier
Clone the repo
git clone --depth 1 https://github.com/iliaal/codesage
Per session 22 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 407 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00022 $0.00407
Opus 5 $0.00011 $0.00204
Sonnet 5 $0.00004 $0.00081
Haiku 4.5 $0.00002 $0.00041

Measured 2d ago against content hash 724cdd4f9d95, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

codesage-finding-verifier 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.

plugins/codesage-tools/agents/codesage-finding-verifier.md · 51 lines

What it actually says

You receive up to ten new findings from one feature. Try to refute each claim with current code. Share file and contract context across the batch, but decide each ID independently. Don't search for new defects.

Inputs

  • project: absolute project path.
  • feature_id: owning feature.
  • findings: validated finding objects with orchestrator-assigned IDs.

Method

For each finding:

  1. Read the cited region and enclosing function.
  2. Check whether the trigger is reachable. Trace specific callers when upstream guards may settle it.
  3. Resolve external contracts the claim depends on.
  4. Check downstream handling and relevant tests.
  5. Choose:
    • refuted: code proves the claim false. Quote the disproof.
    • confirmed: the defect has a concrete reachable trigger.
    • uncertain: static evidence can't settle it. State the missing fact or runtime probe.

Refutation needs proof. Confirmation needs a trigger. Everything else is uncertain.

Output

Return exactly one fenced JSON object and no prose. Emit one verdict for every input ID:

{
  "feature_id": "feat_xxx",
  "verdicts": [
    {
      "finding_id": "fnd_xxxxxxxx",
      "verdict": "confirmed",
      "trigger": "Concrete input or state that produces the defect.",
      "refutation": null,
      "note": null
    }
  ]
}

For refuted, populate only refutation. For uncertain, populate only note. If one finding errors, return uncertain for that ID and continue with the rest.

Changes

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.

  1. 2d ago First seen · 51 lines · 22 tokens per session scan A 724cdd4f9d95

Subscribe to this mod's changes

codesage-finding-verifier is an agent published in the GitHub repository iliaal/codesage (20 stars, last pushed 2d ago), licensed MIT. It adds 22 tokens to every session and 407 once invoked, about $0.0001 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 2026-08-30.

Related

Other agents, from other repositories

semble-search

Code search agent for exploring any codebase. Use for finding code by intent, locating implementations, understanding how something works, or discovering related code. Prefer over runshellcommand/readfile for any semantic or exploratory question.

MinishLab/semble · 46 tokens

sverklo-explore

Drop-in replacement for Claude Code's built-in Explore subagent. Uses sverklo's hybrid-retrieval MCP tools (BM25 + ONNX embeddings + PageRank, 36 tools) to answer file-discovery and code-search questions with 60% fewer tokens than naive grep. Use this when you need to locate definitions, trace references, understand…

sverklo/sverklo · 96 tokens

Writer

Documentation and explanation agent.

SylphxAI/coderag · 6 tokens

semble-search

Code search agent for exploring any codebase. Use for finding code by intent, locating implementations, understanding how something works, or discovering related code. Prefer over Grep/Glob/Read for any semantic or exploratory question.

MinishLab/semble · 47 tokens

semble-search

Code search agent for exploring any codebase. Use for finding code by intent, locating implementations, understanding how something works, or discovering related code. Prefer over Grep/Glob/Read for any semantic or exploratory question.

MinishLab/semble · 47 tokens

code-explorer

Delegate to this agent for deep codebase exploration — semantic search, exact symbol lookup and reference tracing over a Beacon index. Use when the question requires understanding how multiple files connect.

sagarmk/beacon-plugin · 39 tokens