AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Context Budget Review

skill-rlaope-oh-my-hermes-context-budget-review · by rlaope

[omh] Hermes Context Budget Review workflow: plan compact context, token/cost budgets, summarization checkpoints, and overflow recovery before long agent work.

No reviews yet
0 installs
31 views
0.0% view→install

Install

$ agentstack add skill-rlaope-oh-my-hermes-context-budget-review

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-rlaope-oh-my-hermes-context-budget-review)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Context Budget Review? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Context Budget Review

This is a Hermes-native context-budget-review workflow skill.

Why This Exists

context-budget-review ports ECC's context-budget and token-budget instincts into OMH as a compactness gate that protects long-running work without redefining success around a smaller task.

Do Not Use When

  • The user asks for live token/cost telemetry; use ops-observability-card.
  • The user asks to continue a loopable goal; use loop unless budget planning is the explicit blocker.
  • The task is a short one-step answer with no meaningful context risk.

Examples

Good example:

  • Prompt: context-budget-review 이 장기 PR 작업에서 어떤 맥락을 꼭 유지하고 언제 요약해야 하는지 잡아줘.
  • Expected behavior: Prepare contextbudgetplan/v1, mustkeepcontext_pack/v1, checkpoint plan, risk register, and overflow recovery route.
  • Why: The request is about preserving context quality during long-running agent work.

Bad example:

  • Prompt: context-budget-review 토큰 아끼려고 원래 목표를 더 작은 목표로 바꿔줘.
  • Expected behavior: Reject goal shrinking and instead compact context while preserving the full objective and evidence gaps.
  • Why: Budget review optimizes context handling, not the user's requested end state.

Completion Checklist

  • The run or workflow scope, metric window, failure modes, and cost/latency boundary are named.
  • Local telemetry, provider truth, billing truth, and completion evidence are separate states.
  • Warnings name the next measurement or operator review action.

Recovery Notes

  • If provider metrics are unavailable, report only local metadata and mark provider truth not_observed.
  • If cost or latency looks risky, surface a warning plus the next measurement rather than a completion claim.

OMH Context Rail

  • This skill is part of OMH's Hermes workflow layer, not a standalone executor.
  • Product context: OMH is a Hermes-native workflow pack: choose skills, shape work, prepare artifacts, show status, and hand off with evidence boundaries.
  • Current lane: Automation and status (achievements, workspace-audit, production-audit, automation-blueprint, github-event-ops, agent-board, gateway-intent-card, voice-operator, +26 more) - schedules, status, health, and ops review.
  • If the user intent belongs to another OMH lane, hand back to oh-my-hermes or name the adjacent workflow instead of force-fitting this skill.
  • Cross-skill context: every OMH skill: match lane; generic tool can render or execute.
  • Generic-tool checkpoint: image->img-summary; frontend->frontend/a11y/visual-qa; paper->paper-learning; content->content-operator; media->media-input-operator; file->materials-package; search->web-research; live->live-info-operator; audit->workspace/production/security; failures->build-failure; verify->verification-gate; code->codegraph/onboarding/ultraprocess.
  • Coverage: Every generated workflow skill carries this rail.
  • Normal users talk to Hermes; OMH CLI is infra.
  • Boundary: Prepared OMH routing, cards, handoffs, or artifacts are not observed execution, image generation, delivery, review, CI, merge-readiness, or merge evidence.

Use When

Use before long-running research, coding, review, or multi-agent work when context, token, cost, or summary drift could break quality.

Strong routing signals: context-budget-review, context budget review, context budget, token budget review, token budget, prompt budget, context compaction, compact context, too much context, summarization checkpoint, budget this task, 컨텍스트 예산, 토큰 예산, 컨텍스트 압축, 요약 체크포인트

Catalog Metadata

Category: observability Phase: context-budget-review Hermes role: tracker Quality tier: context-budget-gated

Quality bar:

  • Name must-keep context before summarizing or delegating long work.
  • Separate durable requirements, volatile status, file refs, verification evidence, and open blockers.
  • Define checkpoint cadence, overflow recovery, and continuity verification.
  • Use bounded copy while preserving the full objective and evidence gaps.

Handoff policy:

Keep budget design and status narration in Hermes. Provider billing, exact token usage, runtime compaction, and executor cost evidence require observed wrapper, runtime, or provider data.

Required inputs:

  • task or workflow scope
  • expected duration, artifacts, and handoff surfaces
  • available context sources and must-keep facts
  • token, cost, latency, or message-size constraints when known

Expected outputs:

  • contextbudgetplan/v1
  • mustkeepcontext_pack/v1
  • summarizationcheckpointplan/v1
  • budgetriskregister/v1
  • overflowrecoveryroute/v1
  • not-evidence boundary

Artifact expectations:

  • contextbudgetplan/v1 with scope, max visible context, source priority, discard rules, and checkpoint cadence
  • mustkeepcontext_pack/v1 with durable facts, file refs, decisions, PR/CI state, and blocked assumptions
  • summarizationcheckpointplan/v1 with when to compact, what to preserve, and how to verify continuity
  • budgetriskregister/v1 separating estimated cost/token/latency risk from provider-observed truth

Safety rules:

  • Do not claim provider billing, exact token counts, or runtime compaction occurred without observed evidence.
  • Do not drop user requirements, file paths, PR state, verification gaps, or explicit constraints during compaction.
  • Keep estimated budget risk, observed usage, checkpoint summaries, and completion evidence separate.
  • Do not use budget pressure as a reason to shrink the user's requested end state.

Harness Discipline

  • Start from the representative harness registry in oh-my-hermes when the workflow needs coding, research, planning, goal execution, architecture, critique, QA, or documentation lanes.
  • Prefer richer evidence and clearer stop conditions over adding more workflow names.
  • Use specialist lanes only when they change the quality of the answer or verification.

Runtime Evidence

Preferred harness for this skill: context-budget-review.

When local shell access or a bot wrapper is available, record metadata-only evidence:

omh runtime record --skill context-budget-review --harness context-budget-review --status started
omh runtime delegate --run  --requested --not-observed --result not_observed

Record observed delegation results when Hermes or the wrapper exposes them. If delegation is unavailable, keep the result explicit as not_available or not_observed.

Hermes Compatibility Contract

  • Preserve the workflow intent, stop conditions, and verification discipline.
  • Use Hermes-native tools, file operations, and subagent/delegation features when available.
  • Do not require runtime tools, role prompts, or overlays that Hermes Agent does not expose.
  • Respect omh_target_topology/v1 when a wrapper reports it: bind state to the current target/thread, adapt only the parts of this workflow that benefit from multiple Hermes agents, and fall back to single-target behavior when active_agent_count is one.
  • When target topology changes from one to many or many to one, give a concise setup-change comment or use the wrapper's apply action before treating the new topology as persistent.
  • Treat wrapper-supplied memory/context summaries as advisory local context, not proof that opaque Hermes memory was read or changed.
  • When a runtime-specific mechanism appears in imported instructions, translate it to a Hermes-native artifact:
  • goal tools -> .omh/goals/ ledgers, goal_completion_gate/v1, goal_status_card/v1, goal_continuation/v1, or explicit checklists with named next actions,
  • question renderers -> one concise question in the current Hermes interface,
  • native subagents -> Hermes delegation when available, otherwise sequential lanes,
  • shell bridge commands -> optional bridge mode only.

Execution Rules

  1. Load supporting context with skills_list / skill_view when needed.
  2. State the workflow target, constraints, validation evidence, and stop condition.
  3. Keep progress evidence-backed.
  4. Verify with the smallest relevant test or inspection before claiming completion.
  5. If Hermes cannot provide a required runtime capability, say so and use the fallback above.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.