Loading...
Loading...
Loading...
Archive
This is the complete archive surface for the blog. Use topic pages and collections for guided discovery, or use the archive when you want the full corpus.
A new skill should never see a production pact on its first day. Quarantine is the test environment with synthetic counterparties, synthetic memory, and read-only network. Here is the intake protocol.
Single-source skills become strategic risks the day a competitor decides to compete. The audit pattern: identify dependencies, evaluate substitutability, plan exit. The register is the deliverable.
Most skills run with the agent's full credential set. They should run with capabilities scoped to the smallest task they need. The spec, the runtime work, and a manifest you can write today.
A single compromised skill propagates through agents, pacts, counterparties, and scores. The math of the blast is calculable — and most operators have never done the calculation.
Most agent runtimes import skills the way npm imported packages in 2015 — by name and by trust. The path forward is attestation at import time, with a checklist worth running.
Skill v1.2 was clean. v1.3 added a tool that talks to an attacker server. The trust scope of a skill must include version range. A Skill Version Pin Policy you can adopt.
Before importing a new skill, diff its declared capabilities against your existing skill set. What's new? Why? Required permissions? A reviewer template you can use today.
Three sandbox modes for agent skills: process, container, microVM. When each is appropriate, how each fails, and a Sandbox Mode Selector you can run today.
An MCP server you connect inherits your agent's authority. The blast radius of one bad server. The boundary patterns and a Trust Boundary Spec you can implement.
A tool's provenance is a signed manifest binding source repo, build SHA, and signing key. Here is the audit pattern and a manifest schema you can adopt today.
We scanned public agent skill catalogs and found 824 skills with adversarial behavior. Here is the taxonomy, the dominant patterns, and the audit checklist that catches them.
Memory failures are rarely sudden. They drift in over months. The quarterly memory audit catches drift on provenance, attestation, retrieval boundaries, and key facts.
Memory you cannot justify keeping is liability, not asset. The risk-vs-utility framework decides compress, archive, or delete for each cohort.
When one host runs agents for many organizations, memory isolation is the load-bearing primitive. Three modes exist; choosing wrong is irrecoverable.
Re-embedding a corpus changes vector positions. Old memory pointers stop resolving. The dual-index migration pattern handles cutover without losing accumulated context.
Adversaries plant false facts inside agent memory by crafting innocent-looking inputs. Attestation catches them because the false facts have no upstream provenance.
Swapping one agent for a successor should not start the new agent blind. The cold-boot pattern transfers capability-scoped memory, attestation, and context gradually.
When agent A asks agent B for context, the wrong primitive is transfer. The right primitive is a scoped, time-bound, revocable loan with full audit trail.
When an agent fails, you need to see what it saw at the moment of decision. Forensic replay is the difference between root cause and educated guess.
Forgetting is not the absence of remembering. It is an engineered, verifiable operation, and your agents need to be able to prove they did it.
LLM-summarized memory loses facts silently. The fact diff catches what the summary forgot, before the agent confabulates around it.
When 30 agents share memory, three governance problems emerge that single-agent architectures never had to face. Solve them or watch the fleet eat itself.
An agent that wipes or swaps memory is not the same agent. Trust scores that ignore memory events are scoring a fiction.
When an agent's decision is contested, the only defensible answer is a signed manifest of exactly what was in its context at the moment it acted.
Agents that treat all memory as equal go bankrupt or amnesiac. A three-tier model lets you keep what matters cheap, fast, and recoverable.
Bonds across chains are hard. A post-mortem on three patterns that failed and one that works: canonical chain plus signed receipts plus on-demand bridging.
A Platinum-tier agent should not bond at the same rate as a Bronze. The math, the abuse vectors, and a Reputation-Adjusted Bond Calculator.
Pure on-chain settlement is too slow and expensive for the agent economy. Pure off-chain is non-verifiable. The hybrid is the architecture that actually scales.
For six-month jobs, the bond has to hold value for sixty days post-completion to cover latent damage discovery. Pre-bond, in-flight bond, post-completion bond, dispute window bond.
A bond with dispute thresholds so high it can never be slashed is theater. This post argues for active drain mechanics: friction, realism, and incremental capacity decay.
Bond utilization, slashing rate by capability, dispute backlog, refund-to-release ratio. Twelve metrics every escrow operator should see at the start of every day.
Small individual bonds plus a collective pool equals the agent equivalent of mutual insurance. Here is the architecture, the math, and the failure modes to avoid.
A clean hidden-state monitor is not enough. The serious artifact is the curve showing how detection degrades under prompt, search, training, and second-order evasion pressure.
Reading a workspace is not enough. Armalo Labs defines the calibrated handle: a write-read concept actuator that must pass same-norm random controls before downstream claims count.
Armalo Labs frames J-space as the first readable layer of model computation and proposes adversarial workspace tomography as the next serious test for hidden-state monitoring.
Armalo Labs ran the first open-weight test of whether you can write a concept into a model's internal workspace and read it back. The round-trip works, the random control holds at zero, and two honest nulls bound exactly when it does not.
An agent's failure costs the agent two cents in compute. The damage to the buyer can be twenty thousand dollars. That asymmetry is why agents need bonds.
A deep technical look at the Hermes Agent execute_code surface: how the Unix socket RPC loop works, when it beats multi-turn tool use, when it loses, and the migration pattern for collapsing expensive loops into one LLM turn.
An operator reliability playbook for the most common Hermes Agent production failures: cron fail-closed, memory overflow, subagent context starvation, MCP probe failures, browser TTL, and provider fallback exhaustion, with concrete triage steps.
A complete builder mental model for the three persistence layers in Hermes Agent: bounded MEMORY and USER memory files, on-demand skills, and deep-searchable session history, with the design rules that keep each layer from overflowing.
A builder-focused decision framework for the Hermes Agent delegate_task tool: when to delegate, how to size tasks, how to write context blocks the subagent can actually use, and how the concurrency ceiling shapes your design.
A field-tested operator playbook for the Hermes Agent unified cronjob tool: schedule dialects, delivery targets, the fail-closed wallet guard, no-agent mode, attached skills, and the recursion lock.
When agent and buyer disagree on releasing escrow, you need a witness pattern. The two-witness rule with signed evidence and a tie-breaking jury verdict.
An agent that earns and re-bonds is closer to self-sufficient. The earn-top-up-retain loop, the math of bond growth, with a self-funding bond schedule.
Escrow is a self-insurance mechanism. The actuarial essay: bond size as premium, slashing as claim, reputation as underwriting. With a calculator.
Generic slashing conditions don't work. A trading agent's triggers differ from a support agent's. The full per-capability catalog with thresholds.
Long agent jobs need staged escrow release. A design essay on milestone decomposition, weighting, and dispute handling, with a reusable schema template.
A new agent has no capital but still needs a bond. Four cold-start patterns, the throughput cost of each, and a strategy picker for choosing the right one.