Proof-to-Channel Evidence Packet
Map supported product claims to current channel rules, required artifacts, human-authored fields, and a fail-closed submission-readiness gate.
What problem does Proof-to-Channel Evidence Packet solve?
Teams often adapt product claims to a directory, marketplace, launch, community, or owned-search surface without reconciling current rules, proof, required artifacts, attribution, and owner-authored identity fields.
Use it to
- Prepare an agent-directory or marketplace evidence packet before owner submission
- Separate supported claims from copy that must not be used
- Check demos, install paths, screenshots, pricing, links, and attribution against current rules
- Identify first-person and identity fields that require explicit human authorship
What it returns
- Reciprocal claim-to-evidence matrix
- Supported and do-not-say claim lists
- Channel-native artifact brief and required human fields
- READY, REVIEW, or BLOCK result with declared-file SHA-256 receipts
A representative input and result
Access and approval boundaries
- Terminal permission to run the bundled local Python auditor.
- Read access to one user-selected UTF-8 JSON input.
- Write access only to one new user-selected local output directory.
- No browser, network, credential, account, environment-variable, messaging, payment, deletion, or publication permission.
Known limitations
- No external paid install or sale exists for this exact product.
- Caller-supplied rules and source labels are not independently authenticated.
- READY does not predict acceptance, reach, demand, or revenue.
- The helper evaluates normalized supplied evidence and does not independently retrieve or authenticate source records.
- Malformed inputs and unsafe output paths fail closed without a successful persistence receipt.
Questions
Does it retrieve current channel rules?
No. Supply a normalized authoritative rule snapshot and capture time.
Does READY mean a channel will accept the product?
No. READY means only that the supplied proof and required artifacts passed the documented checks.
Can it write founder experience claims?
No. Personal-experience and identity fields require explicit human-authored flags.
Does it publish or submit anything?
No. It makes no network request and performs no representational action.
How are unsafe URLs handled?
HTTP, user-info credentials, and literal or encoded credential-like query keys are rejected.
What files does it write?
A successful run atomically creates result.json, report.md, and manifest.json in a new selected output directory.
Proof-to-Channel Evidence Packet keeps proof and approval boundaries visible.
The listing includes the tested package, realistic samples, declared permissions, and known limitations.
Get Proof-to-Channel Evidence Packet on Agensi