Vendor: Cyan Eclipse Technologies · Platforms: MCP over stdio (HTTP transport on roadmap)
Certification scope
CERTIFIED(local) — Eclipse Bazaar Test Bench 2026-10-26, 21 pass, 1 na of 22 checks (U1 auth: stdio process boundary). Run by W-003 (Verity), suites/fedscope.py. Runs 1–3 FAILED retained append-only (runs 1–2 carried bench-suite defects; run 3 was the pre-fix product card); run 4 is the post-fix re-bench — D1–D3 verified fixed, scoring snapshot byte-identical. N-CHECK: NOVELTY_PROVEN, re-earned in run 4 — deterministic fit + newcomer scoring with every point hand-recomputed by the bench (~58 cases) and traced summaries that name their field basis and flag absent facts instead of inventing them; no fetch or live verb exists on the surface.
What it does
- Read-only search over caller-supplied / bundled LOCAL datasets in SAM.gov-opportunity shape — dataset in, answers out; no live network call and no API key exists in v0.1
- Deterministic fit scoring that always decomposes into named factors with per-factor points and reasons (naics_match 35, set_aside_eligibility 25, deadline_runway 20, past_performance_fit 10, place_of_performance 10) — a score without its breakdown is a defect
- A separate newcomer axis rating realism for a firm with no past performance, from its own five named factors — never folded into fit
- Plain-language summaries built from record fields only: summary_basis names every field a sentence traces to; absent facts become named flags (DEADLINE_UNPARSEABLE / DEADLINE_NAIVE_NO_OFFSET / DEADLINE_PASSED), never invented agency, contact, deadline, or value text
- Fail-closed named refusals throughout (EMPTY_DATASET, PROFILE_REQUIRED, PROFILE_INVALID, per-record RECORD_MALFORMED quarantine — one bad record never kills a search); exactly 4 read-only tools, no fetch/submit/write verb
Tools
opportunity_searchopportunity_scoreopportunity_summarizevalidate_dataset
Problems it addresses
- Search federal contract opportunities with a fit score for my company
- Summarize SAM.gov opportunities in plain language
- Which new opportunities are realistic for a newcomer firm?
Not tested
- Live SAM.gov fetch — NOT built in v0.1 and no API key exists; search runs over caller-supplied / bundled LOCAL datasets in SAM.gov-opportunity shape only
- Hosted / multi-tenant operation — NOT built; local stdio only
Bench scorecard: test-bench/results/eclipse-fedscope-mcp-2026-10-26-4.json
Evaluation-intake listing
This listing records the tool's place in the Bazaar's evaluation
and monitoring pipeline, with the evidence gathered so far printed above — verdict, scope,
and not-tested items included. It is not sold on this site, and its presence in the catalog
is not a claim beyond that evidence. The service behind every verdict is the bench:
what a bench run buys →