DERO MCP Server

Composite: audit a chain artifact (block topoheight, block hash, TX hash, and/or proof string) end-to-end. Returns a verdict (cited_in_false_claim | clean), the actual on-chain facts (block reward, TX acceptance status), an optional proof-string decode, a relayable narrative, and curated rebuttal docs citations.

When to call: when the user asks "what's going on with DERO block X?" / "is this transaction the inflation-claim TX?" / "does this proof string come from a known false claim?" PREFER this over chaining dero_get_block_header_by_topo_height + dero_get_transaction + dero_decode_proof_string yourself: the composite already runs them in parallel, joins them against the flagged false-claim registry, and emits a single verdict field plus a narrative so the agent does not need to compose the rebuttal arc from scratch each time.

Input Requirements (CRITICAL):

• At least ONE of topoheight, block_hash, tx_hash, or proof_string MUST be provided. The composite throws INVALID_INPUT otherwise.

• topoheight is OPTIONAL. Non-negative integer.

• block_hash is OPTIONAL. 64 hex characters.

• tx_hash is OPTIONAL. 64 hex characters.

• proof_string is OPTIONAL. Full deroproof… / DERO bech32 string with HRP.

• include_forge_demo is OPTIONAL (default false). When true AND tx_hash is provided, also forges a fresh demo proof for the same TX (via dero_forge_demo_proof) and embeds it under forge_demo. The demo amount auto-selects: a flagged artifact's pinned amount (e.g. -2.2M for the 2022 claim) > the cited proof_string V > -1 DERO. PREFER setting this true when the agent is fielding a "Verified ✓ means the chain minted coins, right?" question — the embedded forge IS the refutation.

Output: { verdict, inputs, matched_artifacts[], context_note, chain_facts, proof_decode, forge_demo, narrative, related_docs, _diagnostics }. verdict is cited_in_false_claim when any input matches the flagged-artifact registry, else clean. chain_facts is null when no chain-querying input was provided or all daemon calls failed; proof_decode is null when no proof_string was provided. forge_demo is null unless include_forge_demo: true was passed; on success it carries { skipped: false, forged_proof_string, target_amount, ring_slot, ring_size, ring_receiver_address, math, self_check, explorer_display_amount, demo_amount_source } (the slim form — full citations stay at the top level).

PREFER citing the returned related_docs verbatim in the agent response — they are the canonical rebuttal pages and have been validated against the bundled docs index by CI. Quote the context_note when verdict is cited_in_false_claim so the user understands why the artifact matters.

Echo strings through the daemon via DERO.Echo. Useful for round-trip sanity checks.

When to call: when you need to confirm that string payloads reach the daemon intact (e.g. before debugging a malformed call to a more complex tool). PREFER dero_daemon_ping for a lighter-weight liveness probe.

Input Requirements (CRITICAL):

• words MUST be a non-empty array of strings.

Output: the echoed string concatenated by the daemon.

DERO daemon connectivity check via DERO.Ping.

When to call: as the first step in any chain investigation to confirm the daemon is reachable. Call before dero_get_info if you are unsure whether DERO_DAEMON_URL is correctly configured.

Input Requirements: none.

Output: a "Pong" string when the daemon is healthy. On failure this tool returns a structured _meta.error with code RPC_UNREACHABLE and a retry hint.

Decode any DERO bech32 string (dero…, deto…, deroi…, detoi…, or deroproof…) into its constituent parts: HRP, network, compressed public key, and any embedded RPC arguments (CBOR-encoded). For deroproof… strings the "public key" is a derived blinder point used in the proof's commitment math, NOT a wallet pubkey — the tool surfaces is_proof: true so the agent does not mislabel it.

When to call: when the user pastes a deroproof… / integrated-address string and wants to know what value or fields it encodes. PREFER this over chaining bech32 decoders + CBOR libraries yourself: the tool implements the exact same wire format as DEROHE rpc.NewAddress and surfaces the RPC_VALUE_TRANSFER uint64 both as raw and as a signed/wraparound interpretation. The decoder is verified against the publicly-cited 2022 inflation-claim proof string (embedded uint64 = 18446743853709551435 = signed -2,200,000.00181 DERO).

Input Requirements (CRITICAL):

• proof_string is REQUIRED. The full bech32 string including HRP and separator (e.g. deroproof1qyy…). Whitespace is trimmed but the case must be consistent (all lower OR all upper per BIP-0173).

Output: { decoded: { hrp, mainnet, is_proof, public_key_hex, arguments[] }, value_interpretation?: { uint64, signed_int64, is_negative_wraparound, signed_atoms, dero }, context_note?, related_docs? }. arguments is an array of { name, type, type_label, semantic_name?, value }. value_interpretation is present only when an RPC_VALUE_TRANSFER (V) + uint64 (U) argument is found. context_note + extra related_docs are silently attached when the input matches a flagged adversarially-cited artifact. Returns a structured _meta.error with code INVALID_BECH32 on parse failure.

PREFER citing integrity/payload-vs-transaction-proofs and integrity/negative-transfer-protection in any agent response that frames a deroproof… decode result — readers should understand that "this string decodes to value V" is a display-layer fact, not a consensus statement.

Get a single bundled docs page by slug, with plain-text content and headings.

When to call: AFTER dero_docs_search has returned a candidate slug, OR when you have a known slug from a prior citation. PREFER dero_docs_search first when you only have a topic in mind.

Input Requirements (CRITICAL):

• slug MUST be a non-empty doc slug relative to pages/ (e.g. rpc-api/daemon-rpc-api, tutorials/first-app, dero-pay/quick-start).

• product is OPTIONAL but RECOMMENDED to disambiguate identical slugs across docs sites (derod, tela, hologram, deropay).

• offset is OPTIONAL. Long pages (the Captain archive, deep RPC references) are returned in 60000-char chunks; if content_truncated is true in the response, call again with offset: next_offset to fetch the next chunk.

Output: { product, slug, title, headings, content, content_offset, content_length, content_truncated, next_offset, canonical_url, last_updated, source_path }. content_length is the total page size; content_truncated + next_offset signal whether to paginate.

List indexed bundled docs pages across all four products with slugs, titles, and canonical URLs.

When to call: when surveying available docs (e.g. "what TELA tutorials exist?"), OR when you need a slug catalog before invoking dero_docs_get_page. PREFER dero_docs_search when you have a specific question.

Input Requirements:

• product is OPTIONAL. Provide to scope to one of derod | tela | hologram | deropay.

• limit is OPTIONAL (default 120, max 500).

Output: { docs_source, total, products, pages: [{ product, slug, title, canonical_url, last_updated }] }.

Search the bundled DERO documentation index across derod, tela, hologram, and deropay (145+ pages). In-process — no network round trip.

When to call: when you need authoritative docs to answer a DERO question, OR before constructing a citation in your response. Call this BEFORE explaining DVM, RPC methods, TELA contracts, Hologram simulator, or DeroPay webhooks. PREFER returning the top match's canonical_url and slug to the user as a citation.

Input Requirements (CRITICAL):

• query MUST be a non-empty search string.

• product is OPTIONAL. Provide when you know the scope to reduce noise (e.g. tela for TELA-DOC-1 questions).

• section is OPTIONAL. Provide a slug prefix to scope further (e.g. rpc-api under product=derod).

• limit is OPTIONAL (default 8, max 25).

Output: ranked matches with title, slug, headings, excerpt, canonical_url, and score.

Composite: build a fresh deroproof… display object for ANY chosen transaction, ring slot, and amount — including negative amounts that uint64-wrap into the trillions. The forged string is constructed locally from public chain data (no wallet, no keys, no broadcast). On an unpatched explorer it shows Verified ✓ for the chosen amount; on the chain, nothing has changed.

When to call: when a user pastes a deroproof… string and asks "does Verified ✓ mean the chain minted these coins?" Forge an equivalent string for the same TX with a different amount and show the result side-by-side — that is the most direct refutation. Also useful for reproducing the docs/integrity/inflation-claim Part 3 demonstration on arbitrary inputs.

Math: blinder = C[ring_slot] − amount × G, then bech32("deroproof", version || blinder || CBOR({HH: zeros, VU: uint64})). The tool runs the same equation proof.Prove() checks at proof/proof.go:88-95 and self-verifies before returning a string. If the self-check fails, the tool throws rather than emit a string that would not verify.

Input Requirements (CRITICAL):

• Exactly ONE of tx_hash or tx_hex MUST be provided. tx_hash triggers a daemon fetch (and surfaces the receiver address); tx_hex skips the daemon and uses the raw bytes the caller already has.

• ring_slot is OPTIONAL (default 0). Must be in [0, ring_size).

• amount_dero is OPTIONAL (default "-1"). Signed decimal with up to 5 fractional digits, e.g. "-1", "1000000", "-2200000.00181". Negative values produce uint64 wraparounds that unpatched explorers render as positive trillions.

Output: { forged_proof_string, target_amount: { dero, atoms_signed, atoms_uint64 }, ring_slot, ring_size, ring_receiver_address, math: { C_slot_hex, amount_x_G_hex, blinder_hex }, self_check: { verified, method }, explorer_display_amount, context_note, related_docs, _diagnostics }. ring_receiver_address is null when tx_hex was passed (the hex carries publickey pointers, not addresses).

READ-ONLY: this tool never broadcasts, never touches a wallet, never mutates chain state. It computes a string from public inputs and returns it. Annotation readOnlyHint: true is preserved. PREFER citing the returned related_docs (the integrity rebuttal pages) in any agent response — readers should understand the forged string is a display-layer object, not a consensus event.

Fetch a full block (header + miner_tx + transactions + topo position) by height OR hash via DERO.GetBlock.

When to call: when investigating a specific block or verifying a transaction's inclusion. Call dero_get_height first if you do not have a target height. PREFER citing dero_docs_search("block structure") so the user can verify field semantics.

Input Requirements (CRITICAL):

• You MUST provide exactly ONE of hash or height. Providing both or neither returns a structured INVALID_INPUT error.

• hash MUST be exactly 64 hex characters.

• height MUST be a non-negative integer.

Output: full block with block_header, miner_tx, txs, and topo position fields.

Get the total block count via DERO.GetBlockCount. This is a tip count, not a topoheight.

When to call: when you need just the block count (e.g. for delta math against a reference height). PREFER dero_get_height when you need tip and stable heights together.

Input Requirements: none.

Output: { count }.

Get a block header by its 64-char hex hash via DERO.GetBlockHeaderByHash.

When to call: when you have a block hash (e.g. from a tx confirmation) and need its header without the full block body. PREFER dero_get_block when you also need the txs or miner_tx.

Input Requirements (CRITICAL):

• hash MUST be exactly 64 hex characters (matches /^[0-9a-fA-F]{64}$/).

Output: { block_header: {...} }.

Get a block header by topological height (canonical ordering) via DERO.GetBlockHeaderByTopoHeight.

When to call: when you need a header keyed by topo position rather than chain height. Topoheight is the canonical ordering used by DERO indexers; height is the consensus block height.

Input Requirements (CRITICAL):

• topoheight MUST be a non-negative integer no greater than the current topoheight (call dero_get_info first if unsure).

Output: { block_header: { hash, height, topoheight, timestamp, ... } }.

Get a mining block template for a miner payout address via DERO.GetBlockTemplate.

When to call: ONLY when you are actually mining. PREFER dero_get_last_block_header for general chain-tip inspection.

Input Requirements (CRITICAL):

• wallet_address MUST be a valid DERO address (dero1... or deto1...) that will receive the block reward.

• block is OPTIONAL. Set true to include the raw block blob in the response.

• miner is an OPTIONAL label.

Output: block template payload suitable for a mining client. Does NOT submit a block; submission requires the excluded DERO.SubmitBlock method.

Get the ENCRYPTED balance blob for a DERO address at a topo height via DERO.GetEncryptedBalance.

CRITICAL: this returns an opaque encrypted blob, NOT a cleartext balance. Only the wallet holding the spend key can decrypt it. Do NOT present the encrypted bytes as a balance to the user.

When to call: when verifying that an address has on-chain encrypted state (e.g. before attempting a transfer with a wallet you control), or as a sub-step in another tool. PREFER citing dero_docs_search("encrypted balance") so the user understands the opacity.

Input Requirements (CRITICAL):

• address MUST start with dero1 (mainnet) or deto1 (testnet).

• topoheight MUST be an integer; use -1 for the latest chain tip.

• scid is OPTIONAL. Omit for native DERO; provide 64-hex SCID for asset balances.

Output: { status, registration, balance (encrypted blob), ... }.

Estimate gas (compute + storage) for transfers, SC deploys, or SC invokes via DERO.GetGasEstimate. This is a PRE-FLIGHT check; nothing is submitted.

When to call: BEFORE any wallet-side transfer/scinvoke (using external wallet tooling) to size fees, OR when explaining deploy costs to a user. PREFER citing dero_docs_search("gas estimate" or "fees") so the user understands how compute vs storage gas are charged.

Input Requirements (CRITICAL):

• At least ONE of transfers, sc, or sc_rpc MUST be provided.

• sc is the DVM-BASIC contract source string when estimating a deploy.

• sc_rpc is an array of { name, datatype, value } invocation arguments (entrypoint + SC_ID + caller-provided params).

• signer is OPTIONAL but PREFERRED; pass the dero1.../deto1... address that would sign the eventual tx.

Output: { gascompute, gasstorage, status }.

Get the current block heights: tip height, stable height (finalized), and topoheight (canonical ordering) via DERO.GetHeight.

When to call: when you need a quick height snapshot without the full chain-info payload. PREFER dero_get_info when you also need network, version, or difficulty.

Input Requirements: none.

Output: { height, stableheight, topoheight }.

Get DERO daemon and chain metadata: height, topoheight, stableheight, difficulty, version, network, mempool size, and total supply (DERO.GetInfo).

When to call: first thing in any chain-state investigation or sync-health check. Call this BEFORE dero_get_sc, dero_get_transaction, or dero_get_block when you do not already know the current tip. PREFER citing dero_docs_search("DERO.GetInfo") so the user can verify field semantics.

Input Requirements: none.

Output: full chain info JSON including topoheight, stableheight, height, network, version, difficulty, tx_pool_size, and total_supply.

Get the header of the current tip block via DERO.GetLastBlockHeader (no full block body).

When to call: when you need tip block metadata (hash, miner, timestamp, difficulty) without the transactions or miner_tx payload. PREFER dero_get_block when you need transactions or the miner_tx.

Input Requirements: none.

Output: { block_header: { hash, height, topoheight, timestamp, difficulty, ... } }.

Get random registered addresses from the chain (used for ring construction in private transfers) via DERO.GetRandomAddress.

When to call: when building a transfer ring in external wallet tooling, or sampling chain participants. Optional asset SCID limits sampling to holders of that asset.

Input Requirements:

• scid is OPTIONAL. When provided it MUST be exactly 64 hex characters.

Output: { address: string[] }.

Read smart contract state (code and/or stored variables) by SCID via DERO.GetSC. This is the primary entry point for any contract inspection on DERO.

When to call: as the first step in any DVM contract investigation. Pair with dero_docs_search("DVM-BASIC") to interpret the returned code blob. PREFER citing dero_docs_search("smart contract") or dero_docs_get_page on a relevant DVM page so the user can interpret the contract's state model.

Input Requirements (CRITICAL):

• scid MUST be exactly 64 hex characters (the contract id).

• code is OPTIONAL (defaults to true). Set false to skip the source blob when you only need stored variables.

• variables is OPTIONAL (defaults to true). Set false to skip variables when you only need the source.

• topoheight is OPTIONAL. Omit or use -1 for the latest committed state.

Output: { code, balances, variables: { stringkeys, uint64keys }, ... }.

Fetch one or more transactions by hash via DERO.GetTransaction. Each tx is returned with confirmation status, block hash, and (optionally) decoded JSON fields.

When to call: when tracing a tx by hash. Pair with dero_get_sc when the tx invokes a contract. PREFER citing dero_docs_search("transaction structure") so the user can interpret confirmations, ring members, and SC fields.

Input Requirements (CRITICAL):

• txs_hashes MUST be a non-empty array of 64-char hex strings.

• decode_as_json is OPTIONAL. PREFER 1 (any non-zero value) when you want JSON-decoded fields instead of raw blobs.

Output: { txs: [...], txs_as_hex: [...] } with per-tx confirmation, block hash, and (when decoded) parsed payload.

List pending mempool transaction hashes via DERO.GetTxPool.

When to call: when checking unconfirmed activity, watching for a specific tx to land, or estimating mempool pressure. NOTE: tx_hashes may be null or an empty array when the mempool is empty — treat both as "no pending".

Input Requirements: none.

Output: { tx_hashes: string[] | null }.

Resolve a DERO on-chain registered name to its address via DERO.NameToAddress.

When to call: when a user supplies a human-readable name (e.g. "myname") instead of a dero1.../deto1... address.

Input Requirements (CRITICAL):

• name MUST be a non-empty string. Resolution is case-sensitive on the daemon side.

• topoheight MUST be an integer; use -1 for the latest registry state.

Output: { name, address }. On NOT_FOUND the daemon's RPC error is surfaced as a structured _meta.error.

Composite: run a four-step chain (DERO.Ping → DERO.GetInfo → DERO.GetHeight → DERO.GetTxPool) and return a single narrative health report with chain metadata, mempool snapshot, machine-readable signals, and curated docs citations.

When to call: as the first step in any chain-state investigation when the user asks "is the node healthy", "is it synced", or "what is the current state of the chain". PREFER this over chaining the four primitives yourself — the composite handles partial-failure modes and lag-depth classification consistently, and the response already cites the right docs page.

Input Requirements:

• include_tx_pool is OPTIONAL (default true). Set false to skip the mempool snapshot when you only need chain-tip status.

Output: { status, narrative, signals[], chain, mempool, related_docs, _diagnostics }. status is one of healthy | lagging | partial | unreachable. chain is null when DERO.GetInfo was unreachable; mempool is null when skipped or the call failed. On total daemon unreachability the tool returns a structured _meta.error with code RPC_UNREACHABLE.

Composite: send a DVM-BASIC contract source to the daemon's gas estimator, then return the raw estimate alongside a plain-text breakdown (what each gas number means), the parsed contract surface, and curated DVM deploy docs as citations.

When to call: BEFORE asking a wallet to broadcast a deploy transaction, OR when explaining the cost of a contract to a user. PREFER this over chaining dero_get_gas_estimate yourself: this composite already explains gascompute vs gasstorage in plain language, parses the SC source to show what functions the user is about to deploy (reusing extractScSurface from explain_smart_contract), and protects against fabricating a breakdown when the daemon reports 0/0 with a non-OK status.

Input Requirements:

• sc is REQUIRED. The full DVM-BASIC contract source — must contain at least one Function ... End Function block. A function body alone will fail with INVALID_INPUT.

• signer is OPTIONAL. A dero1.../deto1... address that will sign the eventual deploy tx. The daemon uses it for fee context; omitting it still returns a meaningful estimate.

• include_breakdown is OPTIONAL (default true). Set false when you only need the raw numbers (e.g. piping into a fee table).

Output: { estimate: { gascompute, gasstorage, status }, breakdown: { compute_note, storage_note, total_units } | null, signer_used, include_breakdown, sc_surface: { functions, stringkeys, uint64keys, raw_code_length, function_count }, related_docs }. breakdown is null when include_breakdown=false OR when the daemon returned 0/0 with a non-OK status (never fabricated). On DVM compile failure the composite returns a structured _meta.error with code INVALID_INPUT and the daemon's exact compile message in _meta.error.raw.

Composite: fetch a DERO smart contract (code + variables + balances) and return its function surface, a classification of the contract pattern (tela_index | tela_doc | token | registry | minimal | generic), a plain-language narrative, and curated DVM docs citations re-ordered so the most relevant page is first. TELA contracts (apps/files) are detected first and cite the TELA spec; for a deep TELA parse use tela_inspect.

When to call: when the user wants to UNDERSTAND a smart contract — its functions, state shape, or which DVM concept to read about. PREFER this over chaining dero_get_sc with a docs lookup yourself: this composite already parses the DVM-BASIC source for function declarations, sorts stringkeys/uint64keys deterministically, and picks the right docs page from a heuristic so the agent does not have to learn DVM-BASIC syntax to summarize a contract.

Input Requirements:

• scid is REQUIRED. Must be 64 hex chars (the smart contract id). Use 0000…0001 for the on-chain name registry as a known-good example.

• topoheight is OPTIONAL. Provide to inspect the contract at a specific topo height; omit for latest tip.

Output: { scid, topoheight, kind, surface: { functions[], stringkeys[], uint64keys[], balances }, narrative, raw_code_length, has_code, related_docs }. kind is one of tela_index | tela_doc | token | registry | minimal | generic. surface.functions items are { name, args, returns }. has_code is false when the SCID is unknown or has no on-chain code; functions is then [] and the narrative explains the gap. raw_code_length is always present so the agent knows when to fall back to dero_get_sc for the full source.

Composite: take a natural-language intent, fan out parallel scoped searches across the bundled docs for all four DERO products (derod, tela, hologram, deropay), boost any product_hint matches by 1.5×, and return a ranked recommendation list with per-result rationale plus ready-to-cite related_docs.

When to call: at the START of any "where do I read about X?" or "which docs cover Y?" investigation, BEFORE calling dero_docs_search directly. PREFER this over guessing the right product: this composite already runs all four products in parallel, dedupes overlap, surfaces the top heading per result as rationale, and gives you the top-2 citations pre-built. Pass product_hint when the user has already said e.g. "TELA" or "DeroPay" so that product's matches float to the top.

Input Requirements:

• intent is REQUIRED. Free-text description of what the user is trying to do (min 8 chars). Drop verbs and use product nouns like "deploy a TELA app" or "verify a DeroPay webhook signature" for best results.

• product_hint is OPTIONAL. One of derod | tela | hologram | deropay. Multiplies hint-product scores by 1.5×.

• limit_per_product is OPTIONAL (default 2, max 5). Cap per-product hits before merging.

Output: { intent, product_hint, limit_per_product, recommended: [{ product, slug, title, canonical_url, score, boosted_score, rationale }], by_product: { derod | tela | hologram | deropay: { count, top_slug, top_score } }, related_docs: DeroCitation[] }. related_docs is the top-2 picks pre-built as citations the agent can drop straight into a response. On zero matches across every product the composite returns a structured _meta.error with code NO_DOCS_MATCH and a hint to rephrase or drop the product_hint.

Composite: fetch the actual file content stored in a TELA-DOC-1 contract. A DOC's file (HTML/CSS/JS/...) lives inside a DVM-BASIC comment block in the contract code — NOT in a stored variable — so this tool fetches DERO.GetSC, confirms the SCID is a DOC, and extracts the file bytes. Gzip-compressed files (a .gz filename, the TELA-CLI default) are transparently base64-decoded + decompressed to plaintext. Large files paginate via offset.

When to call: when a user wants to READ or inspect the actual code/markup a TELA app file holds (e.g. "show me the HTML of this TELA DOC", "what does this app's app.js contain"). Get DOC SCIDs from tela_inspect on an INDEX first. PREFER this over dero_get_sc: that returns the raw DVM contract wrapper; this extracts just the embedded file content and reports docType, size, and signature presence.

Input Requirements:

• scid is REQUIRED. Must be 64 hex chars and reference a TELA-DOC-1 contract (an INDEX or non-TELA SCID returns INVALID_INPUT with guidance).

• offset is OPTIONAL. Byte offset into the extracted content; pass next_offset to read the next chunk of a large file.

• topoheight is OPTIONAL. Omit for the latest committed state.

Output: { scid, topoheight, filename, doc_type, sub_dir, content_embedded, content, content_offset, content_length, content_truncated, next_offset, compressed, decompressed, stored_filename, signature, signature_note, note, narrative, related_docs }. content is the plaintext file (a 60000-char chunk; paginate via next_offset), or null when content is not embedded (DocShard/STATIC/external). compressed is true for .gz files; decompressed is true when this tool gunzipped them (filename then strips .gz; stored_filename keeps the on-chain name). The contract's author signature presence is reported but NOT cryptographically verified.

Composite: fetch a TELA contract by SCID (DERO.GetSC code + variables) and parse it as either a TELA-INDEX-1 app manifest or a TELA-DOC-1 file contract, auto-detecting which standard it is from the stored keys. TELA is DERO's on-chain web-app platform: an INDEX is the app manifest (like package.json) and DOCs are the individual files (HTML/CSS/JS) stored on chain.

When to call: as the FIRST step whenever a user references a TELA SCID, a .tela dURL app, or asks "what is this TELA contract/app", "what files does this TELA app have", or "is this a TELA INDEX or DOC". PREFER this over dero_get_sc + manual parsing or explain_smart_contract: explain_smart_contract treats TELA contracts as generic DVM and its surface CAPS stored keys at 50, which silently drops DOCn entries on large manifests — tela_inspect reads the raw stringkeys directly so it enumerates ALL DOC references, and it decodes the TELA header/mods/commit schema the generic tool does not understand.

Input Requirements:

• scid is REQUIRED. Must be 64 hex chars (the TELA contract id).

• topoheight is OPTIONAL. Provide to inspect at a specific topo height; omit for the latest committed state.

Output: a discriminated union on kind. tela_index → { scid, topoheight, kind, index: { name, description, icon, durl, mods[], docs:[{position, key, scid, is_entrypoint, malformed}], doc_count, commit, version_history[], current_commit_hash, owner, updateable:'unknown', updateable_note, parse_notes[] }, narrative, related_docs }. tela_doc → { ..., doc: { filename, doc_type, sub_dir, durl, signature, content_embedded, code_size_bytes, immutable }, narrative, related_docs }. not_tela → { ..., kind:'not_tela', reason, observed:{ stringkey_sample[], stringkeys_total, has_code, markers[] }, narrative } — returned (NOT an error) when the SCID is unknown or lacks TELA markers. Updateability cannot be derived from chain state (ringsize is not in GetSC) so it is honestly reported as 'unknown'.

Composite: look up a DERO transaction by hash, classify its confirmation status (confirmed | mempool | unknown) and kind (sc_install | transfer_or_invocation | coinbase | unknown), extract the SC surface inline when the tx is a contract install, and stitch the right DERO tx + DVM docs pages as citations.

When to call: as the FIRST step when investigating any tx by hash — the user asks "what is this tx", "is this confirmed", "what contract did this deploy", or "what does this tx do". PREFER this over chaining dero_get_transaction with dero_get_sc yourself: for SC INSTALL txs the composite already extracts the deployed function surface inline (no second RPC needed because the source is embedded in the tx record), classifies the kind so the agent does not have to inspect the raw shape, and protects against the "empty record" failure mode by surfacing structured TX_NOT_FOUND when the daemon does not know the hash.

Input Requirements:

• tx_hash is REQUIRED. Must be 64 hex chars.

• decode is OPTIONAL (default true). Pass false to ask the daemon to skip the JSON-decoded view (raw hex still comes back; the field hint that the binary is available).

• include_sc_context is OPTIONAL (default true). Set false to skip the inline extractScSurface call for SC install txs (useful when you only need confirmation / ring info).

Output: { tx_hash, confirmation: { status, block_height, valid_block, invalid_blocks, in_pool }, kind, ring: { groups, first_group_size }, reward, signer_visible, native_balance, sc_install: { scid, surface, raw_code_length, has_code } | null, raw_tx_hex_length, narrative, related_docs, _diagnostics }. sc_install is non-null ONLY when the tx is a contract install AND the surface extractor produced something (tx_hash IS the resulting SCID in that case). SC invocation arg decoding is NOT performed — that requires walking the binary tx blob with the DERO tx codec, which is not bundled in this MCP. The composite surfaces raw_tx_hex_length so the agent knows the binary is available via dero_get_transaction. On unknown hash the daemon returns an empty record and the composite returns a structured _meta.error with code TX_NOT_FOUND.