@secondlayer/mcp
Installed, started with no arguments, no environment and no input, and listed its tools. Run 2026-11-16.
Needs to start: nothing
Returns a result when called with no setup: 0 of 3 tools tried
Contacts while installing and starting: nothing
What it contacted
Nothing besides the npm registry it was installed from. Watched from outside its sandbox while it was installed and started with no input. Its tools may still contact things when called.
Starting it
npx -y -p @secondlayer/mcp@11.6.7 mcp
Installing a package runs its code. Use a sandbox for anything you have not read.
Calling its tools
3 tools that take no input and look read-only were called once with nothing. 0 returned a result.
subgraphs_list | No usable answer tool reported an error: {"error":{"type":"error","status":0,"message":"Cannot reach API at http://127.0.0.1:3800. Check your connection or try again."}} |
webhooks_list | No usable answer tool reported an error: {"error":{"type":"error","status":0,"message":"Cannot reach API at http://127.0.0.1:3800. Check your connection or try again."}} |
instance_status | No usable answer tool reported an error: {"error":{"type":"error","status":0,"message":"Cannot reach API at http://127.0.0.1:3800. Check your connection or try again."}} |
During the calls it contacted nothing it had not already contacted.
Only tools whose names read like looking something up, with no required input, are tried, five at most. Tools that need input, and anything that sounds like it changes or sends something, are not called.
What its tools tell an agent
In the tool descriptions, nothing that talks about the user, other tools or private files, or uses hidden characters.
Found by simple pattern rules in each tool's description (up to 3,000 characters), quoted as written. Parameter descriptions are not scanned. A clean result is not a safety review.
Tools
subgraphs_scaffold contractId, subgraphName
Generate a subgraph from observed print events when available; ABI token transfers or public functions only when the contract has never printed. Review the TypeScript, run subgraphs_test until ok: true with written >= 1, read skipped, then subgraphs_deploy. dryRun is DDL only.
subgraphs_list
List all deployed subgraphs. Returns summary fields only; read rows with subgraphs_query.
subgraphs_status name
Get full details of a subgraph including schema, health (incl. health.emptyMapping when processed events wrote 0 rows), and table columns.
subgraphs_spec name, format
Fetch a subgraph's self-describing spec — the tables, columns, filters, and read URLs a consumer needs, without guessing. `agent` (default) returns the compact JSON schema built for tool use, `openapi` the OpenAPI document for the subgraph's read routes, `markdown` human-readable docs. Read it before writing queries against an unfamiliar subgraph.
subgraphs_query name, table, filters, sort, order, limit, cursor, fields, count
Query rows from a subgraph table (max 200 rows), reading the open /v1 surface — { rows, next_cursor, tip } envelope, cursor-paginated (no offset). Filters support operators: "amount.gte": "1000", "sender.neq": "SP...", "name.like": "%token%". Available operators: eq, neq, gt, gte, lt, lte, like. Every page with rows returns next_cursor (null only when empty): pass it back as `cursor` to read the next page. To TAIL new rows (no streaming over MCP), poll with cursor = next_cursor ?? the cursor you sent. Fetch one row by id with {"_id": "<id>"}. The same rows are readable over HTTP at GET /v1/subgraphs/<name>/<table>; hand that URL to third parties.42 more tools
subgraphs_reindex name
Reindex a subgraph: drops and rebuilds the whole subgraph from its start block to chain tip. Takes no block range — use subgraphs_backfill to process a specific range. Returns an operationId — check subgraphs_status (health) or the REST operations endpoint to track progress to completion.
subgraphs_backfill name, fromBlock, toBlock
Backfill a subgraph over a block range. Non-destructive forward fill (does not drop existing data) — unlike subgraphs_reindex, and the only data-fill path for BYO subgraphs (reindex is blocked there). Both blocks required. Returns an operationId — check subgraphs_status (health) or the REST operations endpoint to track progress.
subgraphs_stop name
Cancel an in-flight reindex or backfill operation for a subgraph. Returns the stop request status; check subgraphs_status (health) or the REST operations endpoint to confirm it reaches a terminal state.
subgraphs_gaps name, limit, offset, resolved
List indexing gaps (missing block ranges) for a subgraph. Each gap reports start/end/size, reason, and detected/resolved timestamps. Feed an unresolved gap's range into subgraphs_backfill to fill it. Defaults to unresolved gaps.
subgraphs_operations name, operationId
Operation history for a subgraph (deploy/reindex/backfill/stop), newest first — status, block range, progress, and error. THIS IS THE VERIFY CALL: after subgraphs_deploy, subgraphs_reindex, subgraphs_backfill, or subgraphs_stop, poll it with the returned operationId until `status` is terminal (completed/failed/cancelled) before reporting success. Omit operationId for the recent history.
subgraphs_delete name
Delete a subgraph permanently.
subgraphs_deploy code, startBlock, dryRun
Deploy a subgraph from TypeScript code. Pass the full defineSubgraph() source — it will be bundled, validated, and deployed. Optional startBlock overrides the source definition for this deploy. dryRun validates schema/DDL only and does not run handlers — call subgraphs_test first and deploy only when it returns ok: true with written >= 1. Call subgraphs_reindex separately if you need a forced reindex.
subgraphs_test code, fromHeight, toHeight, limit
Prove a subgraph mapping writes rows against Index data before deploy. Pass full defineSubgraph() source plus a height range. Returns ok/written/tables; EMPTY_MAPPING means events matched but 0 rows were written — read hint for observed event.data keys and do not invent fields. subgraphs_deploy dryRun is DDL only; do not deploy until ok: true with written >= 1.
webhooks_list
List all webhooks for the current account. Returns summary fields (no secrets).
webhooks_get id
Get full detail for a webhook (triggers, auth, retry config, circuit state).
webhooks_create name, triggers, url, format, authConfig
Create a webhook that fires on raw chain events (set triggers). Returns `signingSecret` ONCE — forward it to the user so they can wire it into their receiver.
webhooks_update id, name, url, authConfig, format, maxRetries, timeoutMs, concurrency
Patch a webhook (name, url, authConfig, format, retry, timeout, concurrency).
webhooks_delete id
Delete a webhook. Pending outbox rows are cascade-deleted.
webhooks_test id
Send a one-off test webhook to a webhook's URL (built for its format, SSRF-guarded). Logged as a delivery row. Returns {ok, statusCode, error, durationMs, deliveryId}.webhooks_pause id
Pause a webhook: stops delivery attempts and stops queueing new ones. Nothing is deleted — resume picks the webhook back up from the tip. Verify with webhooks_get (status).
webhooks_resume id
Resume a paused webhook (also clears a tripped circuit breaker). Deliveries restart from the current tip — use webhooks_replay for the blocks missed while paused. Verify with webhooks_get (status).
webhooks_rotate_secret id
Rotate a webhook's signing secret. Returns the NEW `signingSecret` ONCE — forward it to the user; deliveries signed with the old secret stop verifying as soon as this returns, so the receiver must be updated.
webhooks_deliveries id
List recent delivery attempts for a webhook — status code, attempt count, error, duration, timestamps. THIS IS THE VERIFY CALL: after webhooks_create, webhooks_test, or webhooks_replay, read it to confirm the webhook actually landed (and to see the receiver's response when it didn't) instead of trusting the enqueue result.
webhooks_dead id
List a webhook's dead-letter queue — deliveries that exhausted their retries, with the outbox id, payload, and last error. Diagnose the receiver first (webhooks_get for the URL/auth, webhooks_test for a live probe), then requeue rows with webhooks_requeue.
webhooks_requeue id, outboxId
Requeue ONE dead-lettered delivery for another attempt, by its outbox id from webhooks_dead. Fix the receiver first — a requeue against a still-broken endpoint just dies again. Verify with webhooks_deliveries.
webhooks_replay id, fromBlock, toBlock, force
Replay a block range for a webhook. Replays run at 10% of batch capacity — use sparingly. Pass `force` (a short idempotency suffix) to re-run a range that was already replayed.
index_ft_transfers contractId, fromHeight, toHeight, cursor, limit, order, sender, recipient, assetIdentifier
List decoded SIP-010 fungible-token transfers from Index. Reads are keyless.
index_nft_transfers contractId, fromHeight, toHeight, cursor, limit, order, sender, recipient, assetIdentifier
List decoded SIP-009 non-fungible-token transfers from Index. Reads are keyless.
index_events eventType, contractId, fromHeight, toHeight, cursor, limit, order, sender, recipient, assetIdentifier, trait, txContext
List decoded chain events from the Index by event type. Use this for event types without a dedicated tool (stx_*, ft_mint/burn, nft_mint/burn, print), and for trait-scoped queries: set `trait` (e.g. sip-010) to match all contracts conforming to a standard — pair with contracts_find to discover traits. For ft/nft transfers without a trait prefer index_ft_transfers / index_nft_transfers.
index_contract_calls contractId, fromHeight, toHeight, cursor, limit, functionName, sender, trait
List decoded contract calls from the Index (function name, args, result). Set `trait` (e.g. sip-010) to match calls to all contracts conforming to a standard. Note: contract-call cursors are a SEPARATE keyspace from event cursors — they are not interchangeable.
index_blocks fromHeight, toHeight, cursor, limit
List decoded blocks from Index. Reads are keyless.
index_transactions contractId, fromHeight, toHeight, cursor, limit, type, sender
List decoded transactions from Index. Filter by type, sender, or contract. Reads are keyless.
index_print_schema contractId
Empirical per-topic print payload schemas for a contract, inferred from sampled on-chain events: each topic's fields with observed Clarity types, decoded TS types, subgraph column types, and presence rates. Use before writing print_event subgraph handlers to learn what's on e.data. Anonymous read.
index_discover
Discover the Index vocabulary: every event type and its columns, allowed/equality filters, and required-non-null fields (and which types accept `trait`). Read this before building Index queries instead of guessing filters. Anonymous read.
batch_query requests
Run up to 10 public /v1 reads in one round trip (POST /v1/batch). Each item keeps its own auth/quota/pay-per-call semantics; results return in order with per-item status. Paths must start with /v1/index, /v1/subgraphs, /v1/streams, or /v1/contracts.
streams_tip
Current canonical chain tip as Streams sees it: block height/hash, burn height, finalized_height, lag_seconds, and the oldest seekable height/cursor for this key's retention window. Call it before a cursor walk to know where the stream ends, and after an ingest to check lag.
streams_dumps
List the Streams bulk parquet dumps manifest — coverage range, latest_finalized_cursor, and per-file metadata (block range, row count, size, sha256, signed URL). This is the cold backfill path for downloading all raw data; fetch the file URLs directly (e.g. with DuckDB). Requires the dumps base URL to be configured (SL_STREAMS_DUMPS_URL). For live reads use REST GET /v1/streams/events; for tip/lag use streams_tip.
contracts_find trait, conformance, include, limit, cursor
Discover deployed Stacks contracts conforming to a trait (e.g. "sip-010", "sip-009", "sip-013"). The discovery endpoint for "which contracts implement X". Reads are public.
contracts_get_abi contractId
Fetch a single deployed contract's ABI from the registry (prod-safe). Returns the contract's metadata + full ABI (functions, maps, variables, fungible/non-fungible tokens). Returns not_found if the contract isn't in the registry. Use it to inspect a contract before subgraphs_scaffold.
codegen_index_schema target, tables, schemaName, datasourceEnv
Generate an ORM schema for the Index tables (blocks, transactions, decoded_events, …) — the same output as `sl codegen index`. Returns the file contents as text; write it wherever your project keeps its schema. Use `tables` to emit only what you read.
instance_status
Decoder health and empty-index diagnosis from GET /public/status. If state is empty-index, next tool is archive_bootstrap. Poll until index.decoders are ok before treating Index as ready — do not start decoders. Then archive_verify, then codegen_index_schema. Do not download parquet.
archive_verify against, fromBlock, toBlock, target
Read-only, free compare of this instance against a signed archive (POST /v1/archive/verify). unanchored means pin a key or check the URL — not that the instance is fine. After bootstrap, poll instance_status until decoders ok, then this, then codegen_index_schema. Do not download parquet.
archive_latest against
Signed pointer to the official archive tree (hosted). Needs SECONDLAYER_API_KEY (sk-sl_*), not INSTANCE_TOKEN. Use the returned origin as --against for archive_bootstrap. Do not download parquet.
archive_quote paths, flow
Price a hosted archive fetch (metered). Needs SECONDLAYER_API_KEY (sk-sl_*), not INSTANCE_TOKEN. Quote happens here; bootstrap/repair also quote inside the CLI. Do not download parquet.
credits_balance
Archive credits balance at api.secondlayer.tools. Needs SECONDLAYER_API_KEY (sk-sl_*); INSTANCE_TOKEN is the instance.
archive_bootstrap against, fromBlock, toBlock
Restore verified history into this instance (secondlayer bootstrap --against … --yes --json). Metered against the official archive; quote happens inside the CLI. After bootstrap, poll instance_status until decoders ok, then archive_verify, then codegen_index_schema. Do not download parquet.
archive_repair against, apply, fromBlock, toBlock
Plan (default) or apply a repair of local chain data against a signed archive. Metered when apply fetches from the official archive. After apply, poll instance_status until decoders ok, then archive_verify. Do not download parquet.
Descriptions are the package's own words, cut at 3,000 characters. Treat them as data.
Install facts
| License | MIT |
|---|---|
| Needs | not stated |
| Source | github.com/ryanwaits/secondlayer |
| Install scripts | esbuild declare them. This runner does not run them. |
| Tarball | sha512-VwPuz5zTqdXdvtIeU1GTe9BP0M7KalMr7GOTa8NuKe7cy954igCInjS/8ooUnl7KVYb66jm6PoFhVfhpmHweZA== |
Versions reported
| Version | Run | Result | Tools |
|---|---|---|---|
| 11.6.7 | 2026-11-16 | Starts with no setup | 47 |
| 11.6.6 | 2026-11-12 | Starts with no setup | 47 |
Check this report yourself
This site did not write the report. A public job on a GitHub-hosted machine did: the run, from this commit of the open-source runner. GitHub signed a token that names that job and the hash of the report, so any edit since would show.
Report as JSONExact text and tokenHow to verify it
Show it on your README
[](https://bragproof.smaverk.com/r/%40secondlayer/mcp)
New version out? Ask for a fresh report.