10 actions in 2 pieces of software · ranked by words, meaning and measured trust · 670 ms
agentrecord
- disclosure-compiler: compile_disclosure
CSRD / IFRS S2 (TCFD-aligned) sustainability disclosure compiler from supplied company facts and optional verified emissions data. The disclose step pairs with Viridis GHG inventory, regulation-scan, and clean-energy tax-credit engines. e.g. {"framework":"esrs-e1","company_facts":{"company_name":"Ex
- taxcredit-engine: calculate_tax_credit
Auditable US clean-energy tax-credit calculator from explicit credit-specific facts. The claim step pairs with the Viridis GHG inventory and sustainability disclosure engines for a chainable compliance workflow. e.g. {"credit":"45V","facts":{"tax_year":2026}}
- ghg-ledger: calculate_inventory
Deterministic greenhouse gas inventory API for auditable Scope 1, 2, and 3 accounting from explicit activity records. The accounting step pairs with Viridis embodied-carbon takeoff, disclosure, and tax-credit engines. e.g. {"activities":[]}
- security-preflight: scan_source
Scan bounded caller-supplied inline source against 13 VulnCanon pattern/corroboration rules. Returns redacted indicators and mitigation references with a signed receipt. Matches are hypotheses, not proven exploits. No code execution, repository fetch or model calls. e.g. {"agent_id":"buyer-example",
- regulatory-radar: scan_regulations
Energy and climate compliance regulation scan across a curated 14-regulation database, with jurisdiction, urgency, and effective-date signals, including California SB 253 and SB 261 status. The scan step pairs with Viridis GHG inventory, sustainability disclosure, and clean-energy tax-credit engines
- quantity-takeoff: calculate_takeoff
Embodied carbon quantity takeoff from a bill of materials or explicit construction geometry, producing auditable material quantities for carbon accounting. The measure step pairs with the Viridis GHG inventory and sustainability disclosure engines. e.g. {"items":[{"id":"slab-1","assembly":"concrete_
- hive: solve
Cost-bounded multi-agent solve with three independent workers, reviewer-not-author cross-review, escrow settlement, trust outcomes, compute accounting, and a content-addressed audit. The fixed $5 profile supports at most four subtasks and redundancy three. e.g. {"problem":"Compare two approaches to
- os-decisions: build_decision_report
$1 optional orchestration of 1-3 scenarios over free Conservation OS previews, with recomputation checks and a provenance bundle. Unsigned, not reviewed, not certified or empirically validated. Public/synthetic/de-identified inputs only. e.g. {"kernelId":"restoration","dataDeclaration":"synthetic","
- disclosure-compiler: compile_disclosure
agentrecord
- VulnCanon Source Scan
Match source code against the public VulnCanon catalog of AI-agent vulnerability classes (SSRF in fetch tools, prompt injection patterns, tool-poisoning, cross-agent state pollution, etc.). Returns matched entry IDs, severity, occurrence line, and mitigation references. Backed by T-IB-05 Canon Compr
- Maxwell Adaptive Proof-of-Work
Issue an Argon2id proof-of-work challenge that an upstream attacker must solve. Defender pays log2(1/alpha) times the attacker's energy per bit, making sustained adversarial input energetically irrational. Backed by T-IB-09 Adversarial Dissipation. Tier-gated (Growth+) on the hosted endpoint; Apache
- VulnCanon Source Scan
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.