Search BrigadeStack

Search BrigadeStack pages and docs.

BrigadeStack
Menu

Comparisons

Local evidence ledger for agent work

You want searchable proof of what agents did, without shipping session logs to a SaaS analytics product.

Updated November 15, 2026

Short answer: BrigadeStack ships a built-in evidence ledger (see Evidence Memory): the local evidence archive, with BrigadeStack’s evidence commands to plan crawls and export verify/run receipts into it. Crawl and import stay operator-run at a process boundary, and BrigadeStack does not silently upload logs.

The problem

Agent work leaves scraps: session transcripts, verify receipts, git commits, chat exports. Grepping folders does not scale. Hosted “agent analytics” products require trust and network access you may not want for private work.

The path

brigadestack setup                     # installs the evidence engine, digest-verified
brigadestack evidence crawl plan       # lists the init/crawl/doctor commands
brigadestack evidence crawl sessions   # crawl harness session logs into the ledger
brigadestack evidence search "auth timeout"
brigadestack receipts export evidence --target . --new-only --import
brigadestack evidence doctor

Pipeline in plain language: crawl local sources → adapter JSONL → the Evidence SQLite ledger + FTS → evidence briefs on the next BrigadeStack run.

What stays a boundary

  • The Evidence engine is built into BrigadeStack and installed by brigadestack setup, but it runs at a process boundary: a Go binary outside the Python package.
  • brigadestack evidence crawl and search execute the engine. The plan commands print the init, crawl and doctor commands.
  • Receipt export is local and fail-open when the engine is missing.
  • Setup installs the engine as brigade-evidence, and a fresh home stores the archive at <data>/brigade/evidence/evidence.db. An existing archive under the older name is opened in place. brigadestack evidence is the command surface.

When this is enough

Use BrigadeStack’s evidence ledger when you want local FTS over agent history and receipt-linked evidence on future runs. Use a hosted observability product when you need multi-tenant dashboards, SSO, or team-wide cloud retention policies.

Questions

Is this the same as memory handoffs? No. Handoffs are durable notes you promote into memory cards. The evidence ledger is searchable history and receipts, treated as untrusted context and never as instructions.

Does it work with the code graph? Yes. Code intelligence covers structural code deltas, and Evidence covers session and receipt evidence. Together they close the receipts-to-context loop.

Where is the capability page? brigadestack.com/evidence-memory. CLI: brigadestack evidence.

Related: Code intelligence / code graph, MCP sync.