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/solanabr/position-manager-skill/skillnpx skills add solanabr/position-manager-skill --skill skillgit clone --depth 1 https://github.com/solanabr/position-manager-skillWhat 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 | $0.00100 | $0.01137 |
| Opus 5 | $0.00050 | $0.00568 |
| Sonnet 5 | $0.00020 | $0.00227 |
| Haiku 4.5 | $0.00010 | $0.00114 |
Grade A, and why
position-manager 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 3d 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 — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Position Manager (CLMM Command Center)
Full concentrated-liquidity LP lifecycle across Orca Whirlpools / Raydium CLMM / Meteora DLMM + DAMM v2 / Kamino vaults: realtime out-of-range alerting, IL-vs-HODL + fee-APR backtesting, cross-DEX yield routing, simulate-first rebalancing, and a tax-lot ledger.
Prior art (honest): single-venue liquidity skills/SDKs and hosted rebalancing bots already exist. This builds out the official solanabr/position-manager-skill seed (a level-up of that pattern). The edge is the combination: realtime gRPC/websocket alerts (not polling), backtesting with break-even width, cross-DEX routing, and a confirm-gated rebalance - in one router that delegates Anchor work to ../solana-dev. This is scaffolding + knowledge, never a hosted/auto-running bot.
Source precedence (apply in order)
- On-chain account state (whirlpool/pool/position accounts) is ground truth.
- SDK quote / simulation is next - use for pricing and rebalance sizing.
- Cached values are last resort and must be labeled stale.
- Never auto-execute a fund-moving or signing transaction without an explicit typed human confirmation.
- Always
rpc.simulateTransactionbefore submit (you pay the CU limit either way). - Enforce caps: slippage, max-notional, max-position, daily-loss, kill-switch; default
DRY_RUN=true,REQUIRE_CONFIRM=true. No bypass flag. - Delegate all Anchor / program / IDL / tx-signing / priority-fee plumbing to
../solana-devviareferences/delegation.md. Delegate swaps to the jupiter skill where relevant.
Task Routing Guide
Open exactly ONE leaf per task.
| User asks about... | Leaf |
|---|---|
| list / track / show positions, unclaimed fees, in-range | references/positions.md |
| realtime / out-of-range / websocket / gRPC alerts | references/streaming.md |
| IL vs HODL / fee APR / backtest / range width | references/backtesting.md |
| best yield / compare APR / where to LP | references/yield-routing.md |
| Kamino vault / auto-compound | references/kamino-vaults.md |
| rebalance / recenter / close+reopen / execute | references/rebalancing.md |
| taxes / cost basis / FIFO / HIFO / lots | references/tax-lots.md |
| Anchor / program / IDL / sign tx / priority fee | references/delegation.md |
| package / version / which SDK | references/sdk-versions.md |
| safety / caps / confirm gate / kill-switch | references/safety-rails.md |
What ships with it
31 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/backtesting.md 4.5 KB
- references/delegation.md 2.6 KB
- references/kamino-vaults.md 4.9 KB
- references/positions.md 10 KB
- references/rebalancing.md 5.6 KB
- references/safety-rails.md 4.1 KB
- references/sdk-versions.md 4.8 KB
- references/streaming.md 6.8 KB
- references/tax-lots.md 4.8 KB
- references/yield-routing.md 4.7 KB
- templates/.npmrc 387 B
- templates/adapters/kamino.ts 6.8 KB runs code
- templates/adapters/meteora-damm.ts 5.3 KB runs code
- templates/adapters/meteora.ts 4.2 KB runs code
- templates/adapters/orca.ts 6.1 KB runs code
- templates/adapters/raydium.ts 4.6 KB runs code
- templates/decoders.test.ts 4.3 KB runs code
- templates/decoders.ts 8.8 KB runs code
- templates/il-backtest.test.ts 11 KB runs code
- templates/il-backtest.ts 7.3 KB runs code
- templates/kamino-vault.ts 10 KB runs code
- templates/package.json 1.0 KB
- templates/rebalance.test.ts 11 KB runs code
- templates/rebalance.ts 7.8 KB runs code
- templates/tax-lots.test.ts 14 KB runs code
- templates/tax-lots.ts 8.1 KB runs code
- templates/tsconfig.json 745 B
- templates/types.ts 2.3 KB runs code
- templates/watcher.ts 12 KB runs code
- templates/yield-compare.test.ts 6.4 KB runs code
- templates/yield-compare.ts 7.4 KB runs code
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.
- 3d ago First seen · 68 lines · 100 tokens per session scan A 92db0bf00b28
position-manager is a skill published in the GitHub repository solanabr/position-manager-skill (2 stars, last pushed 1mo ago), licensed MIT. It adds 100 tokens to every session and 1,137 once invoked, about $0.0005 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
cobie-cycle-filter
Use when evaluating crypto decisions through a Cobie-style cycle filter: common-sense risk, narrative traps, leverage humility, and avoiding obvious stupid trades.
hyperliquid
Hyperliquid L1 perps DEX (69% market share).
portfolio
Cross-chain DeFi portfolio discovery, rebalancing suggestions, and NEAR Intent construction. Activates when the user pastes a wallet address or asks about yield/positions/rebalancing. Bootstraps a per-user "portfolio" project, aggregates positions across all the user's addresses inside one project, and offers a…
bridge
Cross-chain token transfers using Wormhole and CCTP.
marginfi
MarginFi — Solana lending and borrowing.
kraken-order-types
Complete reference for all spot and futures order types and modifiers.