Audit AI agents, Skills, MCP configurations, tool permissions, and coding-agent setups for prompt injection, excessive agency, credential exposure, data leakage, unsafe code execution, supply-chain risks, memory risks, missing approval controls, and runaway autonomy. Get a calibrated Security Score, Audit Confidence, evidence-backed findings, remediation steps, verification checks, and residual-risk analysis.
# B2B Account Intel Autonomous B2B target account researcher, tech stack detector, buyer pain point mapper, and executive outreach angle generator. Built to run autonomously with **zero paid API keys**.
Turns scattered AI tool intel — tweets, launch posts, screenshots — into a ready-to-use directory database and a publish-ready weekly digest. Give it a niche, get back 15 tools with pricing, use cases, and alternatives, formatted for direct import into Notion, Airtable, or Excel. Or paste this week's findings and get an IG, LinkedIn, or newsletter-ready digest in one pass. Bilingual output, no separate translation pass needed.
Your agent reads user-pasted text, web pages, retrieved docs, or tool output — and you can't tell which sentence an attacker wrote. A one-line "ignore malicious instructions" never held. This delivers a layered injection-defense design: threat model, attack-surface map, input/privilege/output defenses, canary tokens and a classifier, an adversarial red-team set with before/after, and residual risk plus an incident runbook.
Your AI feed says everything happened. It never says what mattered. This Agent searches live across model releases, funding, regulation, research and tooling, then grades each item by signal strength and names who it hits. Every claim carries a source link. Primary sources beat aggregator rewrites. At most two items a day can be graded high signal, and most days that is zero — it says so, instead of inflating a quiet day into a crisis. Ends with what to do this week.
This Skill is built on practical expertise in AI agent safety, prompt-injection defense, source-to-sink privacy review, tool and MCP metadata integrity, memory provenance checks, autonomous action scoping, and evidence-based verification. It helps agents separate instructions from evidence before tool calls, uploads, publishing, deletion, code execution, or network output.
Michelin-kitchen audit for your AI coding agent setup: finds where you are still the bottleneck, places you on the trust ladder, and hands you the one upgrade that lets you run more agents with less supervision.
Audit persistent AI agents for memory contamination, context drift, stale authorization, identity conflicts, and unsafe tool authority. Get evidence-backed findings, minimal corrections, and regression tests.
Tracks official monthly revenue disclosures for 50 Taiwan AI compute and server supply-chain companies across systems, semiconductors and ASIC, memory, PCB and interconnect, power, thermal, and networking. Reports revenue, YoY growth, historical percentile, effective sample size, issuer statements, reporting period, and official sources. No price forecasts or trading advice.