Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/choiyounggi/groundwork/consolidatenpx skills add choiyounggi/groundwork --skill consolidategit clone --depth 1 https://github.com/choiyounggi/groundworkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/choiyounggi/groundwork/consolidate)<a href="https://agentmods.dev/skills/choiyounggi/groundwork/consolidate"><img src="https://agentmods.dev/badge/skills/choiyounggi/groundwork/consolidate.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00065 | $0.01085 |
| Opus 5 | $0.00032 | $0.00543 |
| Sonnet 5 | $0.00013 | $0.00217 |
| Haiku 4.5 | $0.00006 | $0.00109 |
Grade A, and why
consolidate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
memory-loop: consolidate
Capture and expiry keep memory flowing; nothing keeps it coherent. Over many
sessions the index bloats, two files describe the same thing, an old fact
survives next to the newer one that disproved it, and "yesterday" is frozen in a
note whose yesterday is long gone. This skill is the operate-stage cleanup: a
reflective pass that merges long-tier memory into durable, well-organized form —
and, like remember, never writes without your confirmation.
Adapted from xai-org/grok-build's "dream" consolidation pass (Apache-2.0), reframed around memory-loop's save gate and tier model.
When to run
It is a manual skill — there is no forced trigger. Run it when:
MEMORY.mdhas grown large (~120+ lines).- A learning-review nudge surfaced duplicate or contradictory notes.
- A big piece of work finished and left several related memories behind.
Scope — what it may touch
Read MEMORY.md and the tier: long memory files beside it. Exclude:
| Excluded | Why |
|---|---|
Files without a tier key |
Outside the lifecycle (the compatibility contract) — never touch them |
tier: short memories |
The expiry sweep owns these; consolidation leaves them alone |
Identity files (the user/assistant names from setup) |
Continuity anchors — changed only through the identity skill, never here |
Anything already under archived/ |
Already retired |
The five verbs
Review each memory or group through five lenses (the consolidation prompt):
- Merge — fold related information into one coherent, self-contained topic.
- Resolve — on a contradiction, keep only the current truth; a recent session that disproves an older fact wins (note the superseded one as background if it still explains a decision).
- Convert — turn relative dates ("yesterday", "last week") into absolute dates.
- Discard — drop ephemera: greetings, meta-commentary, tool-output noise,
consumed "current state / next steps", and preferences already in global
memory. Discard means move to
archived/, not delete — nothing is lost. - Preserve — keep decisions and their rationale, architecture, preferences, and problem/solution pairs.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 89 lines · 65 tokens per session scan A e16ae501412b
consolidate is a skill published in the GitHub repository choiyounggi/groundwork (3 stars, last pushed yesterday), licensed MIT. It adds 65 tokens to every session and 1,085 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
reflect
Per-project self-improvement - reads the .harness ledger and feedback memories, then proposes gated rule/threshold/ADR changes so the project stops repeating mistakes. Run periodically.
remember
Review auto-memory entries and propose promotions to CLAUDE.md, CLAUDE.local.md, or shared memory. Also detects outdated, conflicting, and duplicate entries across memory layers.
ucai-patterns
Use when the user asks about Claude Code best practices, how to write agents, how to use hooks, how to manage context, or how to work effectively with Claude Code's native systems.
dream
Memory consolidation - review, merge, prune, and index memory files. Run periodically to keep memories organized and up-to-date.
esp32-expert
This skill should be used for ESP32-specific hardware and runtime work in ESP-IDF, Arduino-ESP32, or PlatformIO projects: target/build detection, FreeRTOS tasks and ISRs, memory/DMA, peripherals, power, networking/security, OTA, crash diagnosis, and unattended reliability. Trigger on "debug this ESP32 crash", "review…
audit-context
Measures where a project's Claude context is going before applying any of clean-context's levers. Reports what Glob discovers with and without ignore files applied, the twenty largest files in the repo and which are generated, the cumulative size of every CLAUDE.md loaded for this working directory, and how many MCP…